The tool descriptions pushed agents toward backgrounding anything over
5 seconds, including builds, tests, and installs where you actually
want to wait for the result. This led to unnecessary process_output
round-trips and missed the foreground timeout-to-reattach workflow
entirely.
Reframe background mode as the exception (persistent processes with
no natural exit) and foreground with an appropriate timeout as the
default. Replace "background process" with "tracked process" in
process_output, process_list, and process_signal since they work on
all tracked processes regardless of how they were started.
- Moves `coderd/chatd/`, `coderd/gitsync/`, `enterprise/coderd/chatd/`
under `x/` parent directories to signal instability
- Adds `Experimental:` glue code comments in `coderd/coderd.go`
> 🤖 This PR was created with the help of Coder Agents, and was
reviewed by my human. 🧑💻
- Changes all 41 chat method receivers in `codersdk/chats.go` from
`*Client` to `*ExperimentalClient` to ensure that callers are aware that
these reference potentially unstable `/api/experimental` endpoints.
> 🤖 This PR was created with the help of Coder Agents, and has been
reviewed by my human. 🧑💻
## Problem
When the Coder chat UI is embedded in a VS Code webview, the session
token is set via the Coder-Session-Token header for HTTP requests.
However, browsers cannot attach custom headers to WebSocket connections,
and VS Code Electron webview environment does not support cookies set
via Set-Cookie from iframe origins. This causes all chat WebSocket
connections to fail with authorization errors.
## Solution
Pass the session token as a coder_session_token query parameter on all
chat-related WebSocket connections. The backend already accepts this
parameter (see APITokenFromRequest in coderd/httpmw/apikey.go).
The token is only included when API.getSessionToken() returns a value,
which only happens in the embed bootstrap flow. Normal browser sessions
use cookies and are unaffected.
> Built with [Coder Agents](https://coder.com/agents)
The prompt told the model to "describe the primary intent" and gave
only generic examples, so it stripped PR numbers, repo names, and
other distinguishing details. Added explicit GOOD/BAD examples to
steer away from generic titles like "Review pull request changes".
Also removed "no special characters" which prevented # and / in
identifiers.
- Guard both `onSettled` callbacks in
`archiveAndDeleteMutation.mutate()` with `shouldNavigateAfterArchive()`,
which checks whether the user is still viewing the archived chat (or a
sub-agent of it) before calling `navigate("/agents")`
- Extract `shouldNavigateAfterArchive` into `agentWorkspaceUtils.ts`
with 6 unit test cases covering: direct match, different chat, no active
chat, sub-agent of archived parent, sub-agent of different parent, and
cache-cleared fallback
- Look up the active chat's `root_chat_id` from the per-chat query cache
(stable across WebSocket eviction of sub-agents) to handle the sub-agent
case
> 🤖 This PR was created with the help of Coder Agents, and has been
reviewed by my human. 🧑💻
Bumps ubuntu from `3ba65aa` to `ce4a593`.
[](https://docs.github.com/en/github/managing-security-vulnerabilities/about-dependabot-security-updates#about-compatibility-scores)
Dependabot will resolve any conflicts with this PR as long as you don't
alter it yourself. You can also trigger a rebase manually by commenting
`@dependabot rebase`.
[//]: # (dependabot-automerge-start)
[//]: # (dependabot-automerge-end)
---
<details>
<summary>Dependabot commands and options</summary>
<br />
You can trigger Dependabot actions by commenting on this PR:
- `@dependabot rebase` will rebase this PR
- `@dependabot recreate` will recreate this PR, overwriting any edits
that have been made to it
- `@dependabot show <dependency name> ignore conditions` will show all
of the ignore conditions of the specified dependency
- `@dependabot ignore this major version` will close this PR and stop
Dependabot creating any more for this major version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this minor version` will close this PR and stop
Dependabot creating any more for this minor version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this dependency` will close this PR and stop
Dependabot creating any more for this dependency (unless you reopen the
PR or upgrade to it yourself)
</details>
Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
Bumps rust from `7d37016` to `f7bf1c2`.
[](https://docs.github.com/en/github/managing-security-vulnerabilities/about-dependabot-security-updates#about-compatibility-scores)
Dependabot will resolve any conflicts with this PR as long as you don't
alter it yourself. You can also trigger a rebase manually by commenting
`@dependabot rebase`.
[//]: # (dependabot-automerge-start)
[//]: # (dependabot-automerge-end)
---
<details>
<summary>Dependabot commands and options</summary>
<br />
You can trigger Dependabot actions by commenting on this PR:
- `@dependabot rebase` will rebase this PR
- `@dependabot recreate` will recreate this PR, overwriting any edits
that have been made to it
- `@dependabot show <dependency name> ignore conditions` will show all
of the ignore conditions of the specified dependency
- `@dependabot ignore this major version` will close this PR and stop
Dependabot creating any more for this major version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this minor version` will close this PR and stop
Dependabot creating any more for this minor version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this dependency` will close this PR and stop
Dependabot creating any more for this dependency (unless you reopen the
PR or upgrade to it yourself)
</details>
Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
This PR changes agents desktop resolution from 1366x768 to 1920x1080.
Anthropic requires the that the resolution of desktop screenshots fits
in 1,150,000 total pixels, so we downscale screenshots to 1280x720
before sending them to the LLM provider.
Resolution scaling was already implemented, but our code didn't exercise
it. The resolution bump showed that there were some bugs in the scaling
logic - this PR fixes these bugs too.
The "Task app is not ready to accept input" error occurs when the
agent responds successfully but its status is not "stable" (e.g.
"running"). This is a state conflict, not a gateway error. 502 was
semantically wrong because the gateway communication succeeded.
409 Conflict is correct because the request conflicts with the
agent's current state. This is consistent with how
authAndDoWithTaskAppClient already returns 409 for pending,
initializing, and paused agent states.
## Problem
The `UsageUserDrillIn` play function in
`AgentSettingsPageView.stories.tsx`
flakes in Chromatic (noticed in #23282). After clicking a user row to
drill
into the detail view, sync assertions fire before React finishes the
state
transition — element not found.
<img width="1110" height="649" alt="image"
src="https://github.com/user-attachments/assets/8b5c36c2-09c4-4dd6-a280-ab6379c1464e"
/>
### Root cause
The play function clicks "Alice Liddell" and then waits with
`findByText("Alice Liddell")` before asserting on detail-view content.
But
"Alice Liddell" appears in **both** the list row and the detail header,
so
`findByText` resolves immediately against the stale list-row text that
is
still in the DOM. The same is true for `"@alice"` — `UserRow` renders
`@${user.username}` as a subtitle in the list, and `AvatarData` renders
it
again in the detail view.
### Fix
Gate on `"User ID: ..."` instead — text that **only** renders in the
detail
panel. Once it is in the DOM, the detail view is fully mounted and all
sync
assertions are safe.
Applied to both `UsageUserDrillIn` and `UsageUserDrillInAndBack`, which
had
the same issue.
## Problem
`/agents/analytics` showed an infinite loading spinner. The browser
devtools revealed repeated requests to the chat cost summary endpoint
with `start_date` and `end_date` shifting by a few milliseconds on each
request.
`AgentAnalyticsPage` called `createDateRange(now)` on every render. When
`now` is not passed (production), `createDateRange` falls through to
`dayjs()`, which produces a new millisecond-precision timestamp each
time. Those timestamps became part of the React Query key via
`chatCostSummary()`, so every render created a new query identity, fired
a new fetch, state-updated, re-rendered, and the cycle repeated. The
page never left the loading branch because no query result was ever
observed for the `current` key before it changed.
The same pattern existed in `InsightsContent`, where
`timeRangeToDates()` called `dayjs()` on every render and fed the result
into `prInsights()`.
Storybook didn't catch this because stories pass a fixed `now` prop,
keeping the date range stable.
## Fix
Anchor the date window once using `useState`'s lazy initializer, then
derive `start_date`/`end_date` from the stable anchor during render — no
`useEffect`, no memoization for correctness, just stable input → stable
query key.
- **`AgentAnalyticsPage`**: `const [anchor] = useState<Dayjs>(() =>
dayjs())`, then `createDateRange(now ?? anchor)`. The `now` prop still
takes priority so Storybook snapshots remain deterministic.
- **`InsightsContent`**: Collapses `timeRange` and its anchor into a
single `TimeRangeSelection` state object. A fresh anchor is captured
only when the user changes the selected range (event handler), not on
render. Clicking the already-selected range is a no-op.
Dependabot will resolve any conflicts with this PR as long as you don't
alter it yourself. You can also trigger a rebase manually by commenting
`@dependabot rebase`.
[//]: # (dependabot-automerge-start)
[//]: # (dependabot-automerge-end)
---
<details>
<summary>Dependabot commands and options</summary>
<br />
You can trigger Dependabot actions by commenting on this PR:
- `@dependabot rebase` will rebase this PR
- `@dependabot recreate` will recreate this PR, overwriting any edits
that have been made to it
- `@dependabot show <dependency name> ignore conditions` will show all
of the ignore conditions of the specified dependency
- `@dependabot ignore <dependency name> major version` will close this
group update PR and stop Dependabot creating any more for the specific
dependency's major version (unless you unignore this specific
dependency's major version or upgrade to it yourself)
- `@dependabot ignore <dependency name> minor version` will close this
group update PR and stop Dependabot creating any more for the specific
dependency's minor version (unless you unignore this specific
dependency's minor version or upgrade to it yourself)
- `@dependabot ignore <dependency name>` will close this group update PR
and stop Dependabot creating any more for the specific dependency
(unless you unignore this specific dependency or upgrade to it yourself)
- `@dependabot unignore <dependency name>` will remove all of the ignore
conditions of the specified dependency
- `@dependabot unignore <dependency name> <ignore condition>` will
remove the ignore condition of the specified dependency and ignore
conditions
</details>
Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
Continuation of https://github.com/coder/coder/pull/23067
Add filtering to the paginated org member endpoint (pretty much the same
as what I did in the previous PR with group members, except there I also
had to add pagination since it was missing).
## What
Replace the hardcoded 30-day date window on the Agents Settings Usage
page (`/agents/settings/usage`) with an interactive date-range picker.
## Why
The usage page previously showed a static 30-day lookback with no way
for admins to adjust the time window. The backend API already supports
`start_date`/`end_date` parameters — only the frontend was missing the
controls.
## How
- Reuse the existing `DateRange` picker component from Template Insights
- Store selected dates in URL search params (`startDate`/`endDate`) for
persistence across navigation
- Default to last 30 days when no params are present
- Memoize date range for stable React Query keys
- Both the user list and per-user drill-in views respect the selected
range
- Normalize exclusive end-date boundaries for display
- Preset clicks (Last 7 days, etc.) apply immediately with a single
click
- Semi-transparent loading overlay during data refetch
## Changes
- `site/src/pages/AgentsPage/SettingsPageContent.tsx` — Replace
hardcoded range with interactive picker, URL param state, memoized
params, refetch overlay
- `site/src/pages/AgentsPage/SettingsPageContent.stories.tsx` — Add
stories for date filter interaction, preset single-click, and refetch
overlay
- `site/src/pages/TemplatePage/TemplateInsightsPage/DateRange.tsx` —
Detect preset clicks and apply immediately (single-click) instead of
requiring two clicks
## Validation
- TypeScript ✅
- Biome lint ✅
- Storybook tests 13/13 ✅
- Visual verification via Storybook ✅
The attribution footer in the PR style guide assumed all AI-generated
PRs come from Claude Code using Claude Sonnet 4.5. PRs can be generated
through different tools and models (e.g. Coder Agents), so a hardcoded
template is misleading.
Co-authored-by: Michael Suchacz <ibetitsmike@users.noreply.github.com>
## Summary
Adds inline editing of existing per-user and per-group chat usage limit
overrides from the Limits tab. Admins can now click Edit on any override
row to modify the spend limit in-place, using the same form used for
adding overrides.
## Changes
**Backend** (`coderd/chats_test.go`)
- Added `UpdateUserOverride` and `UpdateGroupOverride` test cases
covering the upsert-in-place behavior.
**Frontend** (3 component files + 2 story files)
- `LimitsTab.tsx`: Edit state management, mutual-exclusion between
user/group edit modes, and handlers that prefill the form from the
existing override.
- `GroupLimitsSection.tsx`: Edit button per row, read-only group
identity in edit mode, Save/Cancel actions, disable states during
pending operations.
- `UserOverridesSection.tsx`: Same pattern as groups — Edit button,
read-only user identity, Save/Cancel, proper disable states.
- New Storybook stories for both sections (Default, EmptyState,
AddForm, EditForm).
## UX behavior
- Clicking Edit opens the inline form with the current spend limit
prefilled and the entity shown as read-only.
- Save uses the existing PUT upsert endpoint (no new API surface).
- Cancel returns to normal list view with form state cleared.
- Edit modes are mutually exclusive — editing a user override closes
any open group form and vice versa.
- All buttons and inputs disable during pending mutations.
- Add and delete continue to work after editing.
Partially addresses #21813 (still need to make changes to the "add user"
button to be complete)
Since there are a lot of user tests already, I moved them into
`coderdtest` to be shared.
We currently call GetWorkspaceAgentByID millions of times at scale
unnecessarily. This PR embeds immutable fields into the relevant
services instead of fetching for them every time.
resolves https://github.com/coder/scaletest/issues/84
Confirmed with a 10k scaletest that this changeset takes the query from
10M+ queries down to 39k
Add Unwrap() to StatusWriter so http.ResponseController.SetWriteDeadline
can reach the underlying net.Conn through the middleware wrapper. Without
this, the agent's 20s WriteTimeout killed blocking process output
connections.
Also add 30s headroom to the write deadline in handleProcessOutput so
the response can be written after a full-duration blocking wait.
On the tool layer, waitForProcess and the process_output tool now try a
non-blocking snapshot on any error, not just context timeout. Transport
errors (like the WriteTimeout EOF) previously returned with no process
ID and no recovery path. Now if the process finished, the result is
returned transparently. If still running, the error includes the process
ID and tells the agent to use process_output.
The 5ms ServiceBannerRefreshInterval caused excessive DRPC
connection churn (200 calls/s) under the race detector, creating
heavy mutex contention on FakeAgentAPI and significant CPU overhead.
This made the test timing-sensitive in ways that manifested as
session.Wait() hangs, killing the test binary via timeout.
Three changes:
- Increase refresh interval from 5ms to testutil.IntervalFast (25ms),
reducing DRPC connection churn and mutex contention by 5x.
- Replace bare <-ready receives with testutil.TryReceive so the test
fails with context expiry instead of hanging indefinitely.
- Add a timeout to session.Wait() in testSessionOutput to prevent any
SSH session hang from killing the entire test binary.
Fixescoder/internal#1417
Replace afero.TempFile (which uses os.CreateTemp with mode 0600)
with a custom createTempFile that uses OpenFile with mode 0666.
This lets the kernel apply the process umask, matching the default
behavior of os.Create. New files now get ~0644 (with standard
umask) instead of 0600.
Extract atomicWrite(ctx, path, mode, haveMode, reader) to share
the entire temp-file lifecycle between writeFile and editFile.
## Problem
Flaky e2e test `create workspace and overwrite default parameters` — the
boolean parameter verification reads `"true"` when it should be
`"false"`.
`verifyParameters` in `site/e2e/helpers.ts` used a one-shot
`isChecked()` for boolean parameters (line 214), while the
`string`/`number` path used Playwright's auto-retrying `toHaveValue()`
with a 15-second timeout. When the settings/parameters page hydrates
with React Query data, the Switch can briefly render the default value
(`true`) before settling on the override (`false`). The one-shot check
captures the stale state.
## Fix
Replace the one-shot `isChecked()` + `expect().toEqual()` with
Playwright's auto-retrying `toBeChecked()` / `not.toBeChecked()`
assertions using a 15-second timeout, matching the pattern already used
for string/number parameters.
Fixescoder/internal#1414
Authored by coder agent 🤖
---------
Co-authored-by: Copilot Autofix powered by AI <175728472+Copilot@users.noreply.github.com>
- Add `agents_workspace_ttl` site config (default: whatever the template
says a.k.a. `0s`)
- Expose via GET/PUT `/api/experimental/chats/config/workspace-ttl`
- Chat tool reads setting and passes `TTLMillis` on workspace creation
- Existing autostop infrastructure handles the rest (zero changes to
LifecycleExecutor, CalculateAutostop, or activity bumping)
- ⚠️ Template-level `UserAutostopEnabled=false` overrides this global
default. Not touching this.
- Frontend: "Workspace Lifetime" control in /agents/settings Behavior
tab (admin-only)
> This PR was created with the help of Coder Agents, and has been
reviewed by several humans and robots. 🤖🤝🧑💻
## Summary
Adds a new `coderd/chatd/mcpclient` package that connects to
admin-configured MCP servers and wraps their tools as
`fantasy.AgentTool` values that the chat loop can invoke.
## What changed
### New: `coderd/chatd/mcpclient/mcpclient.go`
The core package with a single entry point:
```go
func ConnectAll(
ctx context.Context,
logger slog.Logger,
configs []database.MCPServerConfig,
tokens []database.MCPServerUserToken,
) (tools []fantasy.AgentTool, cleanup func(), err error)
```
This:
1. Connects to each enabled MCP server using `mark3labs/mcp-go`
(streamable HTTP or SSE transport)
2. Discovers tools via the MCP `tools/list` method
3. Wraps each tool as a `fantasy.AgentTool` with namespaced name
(`serverslug__toolname`)
4. Applies tool allow/deny list filtering from the server config
5. Handles auth: OAuth2 bearer tokens, API keys, and custom headers
6. Skips broken servers with a warning (10s connect timeout per server)
7. Returns a cleanup function to close all MCP connections
### Modified: `coderd/chatd/chatd.go`
In `runChat()`, after loading the model/messages but before assembling
the tool list:
- Reads `chat.MCPServerIDs` from the chat record
- Loads the MCP server configs from the database
- Resolves the user's auth tokens
- Calls `mcpclient.ConnectAll()` to connect and discover tools
- Appends the MCP tools to the chat's tool set
- Defers cleanup to close connections when the chat turn ends
The chat loop (`chatloop.Run`) already handles tools generically —
MCP-backed tools are invoked identically to built-in workspace tools. No
changes needed in `chatloop/`.
### New: `coderd/chatd/mcpclient/mcpclient_test.go`
10 tests covering:
- Tool discovery and namespacing
- Tool call forwarding and result conversion
- Allow/deny list filtering
- Connection failure handling (graceful skip)
- Multi-server support with correct prefixes
- OAuth2 auth header injection
- Disabled server skipping
- Invalid input handling
- Tool info parameter propagation
## Design decisions
- **Tool namespacing**: `slug__toolname` with double underscore
separator. Avoids collisions with tools containing single underscores.
Stripped when forwarding to `tools/call`.
- **Connection lifecycle**: Fresh connections per chat turn, closed via
`defer`. Matches the `turnWorkspaceContext` pattern.
- **Failure isolation**: Each server connects independently. A broken
server doesn't fail the chat — its tools are simply unavailable.
- **No chatloop changes**: The existing `[]fantasy.AgentTool` interface
is already fully generic.
## What's NOT in this PR (follow-ups)
- Frontend MCP server picker UI (selecting servers for a chat)
- System prompt additions describing available MCP tools
- Token refresh on expiry mid-chat
- The deprecated `aibridged` MCP proxy cleanup
The git hooks now classify staged files and select either the full
or lightweight make target. This was missing from the contributing
guide after #23358 landed.
Also add actionlint config to suppress a pre-existing SC2016 false
positive in the triage workflow. Shellcheck disable directives
don't work inside heredocs when actionlint drives shellcheck.
Downgrades the "reporting script completed" log in `agentscripts` from
ERROR to WARN.
During agent reconnects, the `scriptCompleted` RPC can race with the
connection teardown, producing a "connection closed" error. Since
`slogtest` treats ERROR logs as test failures, this causes
`TestAgent_ReconnectNoLifecycleReemit` to flake on macOS.
A failed timing report is non-fatal — the script itself has already
finished, and the agent will continue operating normally. WARN is the
appropriate severity, consistent with the call site in
`agent.go:createDevcontainer`.
Also switches from `fmt.Sprintf` to structured `slog.Error` fields for
consistency with the rest of the codebase.
Fixescoder/internal#1410
> **PR Stack**
> 1. #23351 ← `#23282`
> 2. #23282 ← `#23275`
> 3. #23275 ← `#23349`
> 4. **#23349** ← `main` *(you are here)*
---
Retry events were published only to the local in-process stream via
`publishEvent()`. When pubsub is active, `Subscribe()`'s merge loop only
forwarded durable events (messages, status, errors) from pubsub
notifications,
so retry events were silently dropped for cross-replica subscribers.
This adds a `publishRetry()` helper that publishes both locally and via
pubsub,
and extends the `Subscribe()` notification handler to forward retry
events.
**Changes:**
- `coderd/pubsub/chatstreamnotify.go`: add `Retry` field to notify
message
- `coderd/chatd/chatd.go`: add `publishRetry()`, update `OnRetry`
callback,
extend `Subscribe()` to forward `notify.Retry`
- `coderd/chatd/chatd_internal_test.go`: focused pubsub delivery test
- `enterprise/coderd/chatd/chatd_test.go`: cross-replica end-to-end test
Audited exported helpers in `coderd/util/*`, `testutil`, `cryptorand`,
and friends, then replaced duplicated implementations with canonical
versions.
- **fix: `maps.SortedKeys` generic signature** — value type was
hardcoded to `any`, making it impossible to actually call. Added second
type parameter `V any`. Added table-driven tests with `cmp.Diff`.
- **refactor: replace ad-hoc ptr helpers with `ptr.Ref`** — removed
`int64Ptr`, `stringPtr`, `boolPtr`, `i64ptr`, `strPtr`, `PtrInt32`
across 6 files.
- **refactor: replace local `sortedKeys`/`sortKeys` with
`maps.SortedKeys`** — now that the signature is fixed, scripts can use
it.
- **refactor: replace hand-rolled `capitalize` with
`strings.Capitalize`** — the typegen version was also not UTF-8 safe.
> 🤖 This PR was created with the help of Coder Agents, and was reviewed
by my human. 🧑💻
Pre-commit classifies staged files and runs make pre-commit-light
when no Go, TypeScript, or Makefile changes are present. This
skips gen, lint/go, lint/ts, fmt/go, fmt/ts, and the binary
build. A markdown-only commit takes seconds instead of minutes.
Pre-push uses the same heuristic: if only light files changed
(docs, shell, terraform, etc.), tests are skipped entirely.
Falls back to the full make targets when Go/TS/Makefile changes
are detected, CODER_HOOK_RUN_ALL=1 is set, or the diff range
can't be determined.
Also adds test-storybook to make pre-push (vitest with the
storybook project in Playwright browser mode).
- Remove `TRIVY_VERSION` ARG and trivy CLI install block from
`dogfood/coder/Dockerfile`
- The `trivy` job in `.github/workflows/security.yaml` is kept — it uses
`aquasecurity/trivy-action` pinned to a known-good commit
> 🤖 This PR was created with the help of Coder Agents, and was reviewed
by my human. 🧑💻
Adds a repo-local agent skill at `.agents/skills/pull-requests/SKILL.md`
that guides agents through the PR lifecycle for this repository:
creating, updating, and following up on pull requests.
Covers lifecycle rules (reuse existing PRs, default to draft), local
validation commands (`make pre-commit`, `make lint`, etc.), PR
title/description conventions, CI check follow-up, and explicit
guardrails against common mistakes.
## Problem
Deleting a chat provider that has models referenced by existing chats
returns a raw PostgreSQL foreign key violation error to the user:
```
pq: update or delete on table "chat_model_configs" violates foreign key
constraint "chat_messages_model_config_id_fkey" on table "chat_messages"
```
This happens because `DELETE FROM chat_providers` cascades to
hard-delete
`chat_model_configs` rows, but `chat_messages` and `chats` still
reference
them with the default `RESTRICT` behavior.
## Fix
Check for `IsForeignKeyViolation` on the two relevant constraints and
return a 400 Bad Request with `"Provider models are still referenced by
existing chats."`, matching the existing FK error handling pattern used
elsewhere in the same file.
The Base URL field in the provider config form always showed
`https://api.example.com/v1` as its placeholder, regardless of the
selected provider. This was confusing — I added `/v1` to my Anthropic
base URL because the placeholder suggested it, but the Anthropic SDK
already prefixes its request paths with `v1/`, so this doubled it up
and broke requests.
The placeholder is now provider-aware:
- **Anthropic, Bedrock, Google** → `https://api.example.com`
- **OpenAI-family providers** (openai, openai-compat, openrouter,
vercel, azure) → `https://api.example.com/v1`
Replace the 200ms polling loop in chatd's execute and
process_output tools with server-side blocking via sync.Cond
on HeadTailBuffer.
The agent's GET /{id}/output endpoint accepts ?wait=true to
block until the process exits or a 5-minute server cap expires.
The process_output tool blocks by default for 10s (overridable
via wait_timeout), and falls back to a non-blocking snapshot on
timeout. The execute tool's foreground path makes a single
blocking call instead of polling.
Related #23316
Adds a warning comment to dbauthz.AsSystemRestricted to hopefully nudge agents away from it.
---------
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
- Rate-limit "chat stream buffer full" and "dropping chat stream event"
WARN logs to at most once per 10s per chat.
- Intermediate drops not logged; WARN includes `dropped_count`.
- Per-chat tracking on `chatStreamState` using timestamp comparison
against `quartz.Clock` — no global tickers, no new `Server` fields.
- Subscriber and buffer drop counters reset at all lifecycle boundaries.
> 🤖 This PR was created with the help of Coder Agents, and was reviewed
by my human. 🧑💻
## Description
Blocks `CONNECT` tunnels to private and reserved IP ranges in
aibridgeproxyd, preventing the proxy from being used to reach internal
networks.
The Coder access URL is always exempt (hostname+port match) so the proxy
can reach its own deployment. It is possible to exempt additional ranges
via `CODER_AIBRIDGE_PROXY_ALLOWED_PRIVATE_CIDRS`.
DNS rebinding is handled differently per path:
* Direct (no upstream proxy): validate the resolved IP right before the
TCP dial, no window between check and connect.
* Upstream proxy: Resolves and checks before forwarding to the upstream
dialer. A small rebinding window exists since the upstream proxy
re-resolves independently.
## Changes
* Add blocked IP denylist covering private, reserved, and
special-purpose ranges
* Add `AllowedPrivateCIDRs` option with CLI flag and env var
* Wire IP checks into `proxy.ConnectDial` for both upstream and direct
paths
* Add tests for blocked/allowed cases across direct dial, upstream
proxy, CIDR exemptions, and CoderAccessURL exemption
Notes: documentation will be handled in a follow-up PR.
Closes: https://github.com/coder/security/issues/124
- Replace real healthcheck with mock `HealthcheckFunc` that returns a
canned report instantly
- Remove healthcheck cache-seeding goroutine/channel workaround
- Remove `HealthcheckTimeout: testutil.WaitSuperLong` (no longer needed)
- Reduce `setupCtx` from `WaitSuperLong` (60s) to `WaitLong` (25s)
The DERP healthcheck performs real network operations (portmapper
gateway probing, STUN) that hang for 60s+ on macOS CI runners. Since
`TestSupportBundle` validates bundle generation, not healthcheck
correctness, a canned report eliminates this entire class of flake.
Fixescoder/internal#272
> 🤖 This PR was created with the help of Coder Agents, and was reviewed
by my human. 🧑💻
handleProcessOutput and handleSignalProcess did not check the
chat ID from the request. Any caller that knew a process ID
could read output or signal processes belonging to other chats.
handleListProcesses already filtered by chat ID. Apply the
same check to the output and signal handlers. Non-chat callers
(no Coder-Chat-Id header) are allowed through for backwards
compatibility.
Both writeFile and editFile now use the same atomic write strategy:
temp file in the same directory, write, rename. This ensures a
failed write leaves the original file intact instead of truncated.
editFile already used temp-and-rename but lost the original file's
permissions because afero.TempFile creates with mode 0600. Both
functions now Chmod after rename to preserve the original mode.
writeFile also swallowed io.Copy errors (logged but returned HTTP
200). Fixed to return the error so the client knows the write
failed.
## Summary
The search field on `/agents/settings/usage` previously only matched
against usernames. This updates the SQL query to also match against the
user's display name via `ILIKE`, and updates the frontend placeholder
and variable names to reflect the broader search scope.
## Changes
- **SQL** (`coderd/database/queries/chats.sql`,
`coderd/database/queries.sql.go`): Added `OR u.name ILIKE '%' ||
@username::text || '%'` to the `GetChatCostPerUser` query's WHERE
clause.
- **Frontend** (`site/src/pages/AgentsPage/SettingsPageContent.tsx`):
Renamed `usernameFilter`/`debouncedUsername` to
`searchFilter`/`debouncedSearch`, updated placeholder to "Search by name
or username".
---
PR generated with Coder Agents
The ampersand detection treated bash's pipe-stderr operator (|&)
as a trailing & for backgrounding, stripping it and producing a
broken pipe command. Also adds tests for execute.go and chatloop
context limit helpers, covering previously untested edge cases.
- Wire `workspacestats.ActivityBumpWorkspace` into `trackWorkspaceUsage`
so the workspace build deadline is extended each time the chat heartbeat
fires
- Prevents mid-conversation autostop for chat workspaces
- Updates `TestHeartbeatBumpsWorkspaceUsage` verifying the deadline bump
> This PR was created with the help of Coder Agents, and was reviewed by two humans and their pet robots 🧑💻🤝🤖
Still not applying at the dynamic parameters websocket. The wsbuilder is
the source of truth for previous values, so this is the most accurate
and still will fail in the synchronous api call to build a workspace.
This mirrors how we handle immutable params.
Closes https://github.com/coder/coder/issues/19064
This adds the UI but does not add it to the Settings sidebar. Until it's
actually functional and usable (which will come in future PRs) it will
remain hidden.
Next step is wiring this up to chats and actually testing the full flow
end-to-end, but we aren't there yet.
- Replace fixed `100vh` spacer at bottom of diff list with a dynamically
sized one
- Adds "X files changed" message to the bottom of the diff
> This PR was created with the help of Coder Agents, and was reviewed by several humans and robots. 🧑💻🤝🤖
There are 333 stories with play functions but no local way to run them.
CI uses Chromatic, which means broken play functions aren't caught until
after push. For agents, the feedback loop is even worse since they can't
open a browser.
This adds the `@storybook/addon-vitest` integration so play functions
can run locally via vitest + Playwright:
```sh
pnpm test:storybook
pnpm test:storybook src/path/to/component.stories.tsx
```
The vitest config is restructured into two projects (`unit` and
`storybook`).
The processChat defer at line 2464 catches panics on its main
goroutine and transitions the chat to error status. This was
previously untested.
The test wraps the database Store to panic during PersistStep's
InTx call, which runs synchronously on the processChat goroutine.
A tool-level panic wouldn't work because executeTools has its own
recover that converts panics into tool error results.
scheduleStreamReset() fired for every durable message event with
changed=true, including user messages (e.g. promoted queued messages).
When a batch contained trailing message_parts followed by a user
message event, the batch loop flushed the parts (building stream
state), then scheduleStreamReset cleared it immediately.
Restrict the reset to assistant messages, which are the only role
that ends a streaming turn.
Dynamic parameters supports ephemeral parameters. Updated the test to
use dynamic parameters.
Ephemeral params **require** a default value.
Closes https://github.com/coder/coder/issues/19065
## Problem
Anthropic rejects requests containing empty text content blocks with:
```
messages: text content blocks must be non-empty
```
Empty text parts (`""` or whitespace-only like `" "`) get persisted in
the database when a stream sends `TextStart`/`TextEnd` with no
`TextDelta` in between. On the next turn, these parts are loaded from
the DB and sent to Anthropic, which rejects them.
## Fix
Filter empty/whitespace-only text and reasoning parts at the two LLM
dispatch boundaries, without modifying persistence (the raw record is
preserved):
- **`partsToMessageParts()`** in `chatprompt.go` — filters when
converting persisted DB messages to fantasy message parts for LLM calls.
This is the last gateway before the Anthropic provider creates
`TextBlockParam` objects.
- **`toResponseMessages()`** in `chatloop.go` — filters when building
in-flight conversation messages between steps within a single turn.
Note: `flushActiveState()` (the interruption path) already had this
guard — the normal `TextEnd` streaming path did not, but since we're not
changing persistence, the fix is applied at the dispatch layer.
## Problem
When `Store: true` is set for OpenAI Responses API calls (the new
default), multi-turn conversations with reasoning models fail on the
second message:
```
stream response: bad request: Item 'rs_xxx' of type 'reasoning' was provided
without its required following item.
```
The fantasy library was reconstructing full `OfReasoning` input items
(with encrypted content and summary) when replaying assistant messages.
The API cannot pair these reconstructed reasoning items with the output
items that originally followed them because the output items are sent as
plain `OfMessage` without server-side IDs.
## Fix
Updates the fantasy dependency (`kylecarbs/fantasy@cj/go1.25`) to skip
reasoning parts during conversation replay in `toResponsesPrompt`. With
`Store` enabled, the API already has the reasoning persisted server-side
— it doesn't need to be replayed in the input.
Fantasy PR: https://github.com/charmbracelet/fantasy/pull/181
## Testing
Adds `TestOpenAIReasoningRoundTrip` integration test that:
1. Sends a query to `o4-mini` (reasoning model with `Store: true`)
2. Verifies reasoning content is persisted
3. Sends a follow-up message — this was the failing step
4. Verifies the follow-up completes successfully
Requires `OPENAI_API_KEY` env var to run.
Omit rehype-raw from the Streamdown rehype plugin list so
HTML-like syntax in LLM output is escaped as text instead of
being parsed by the HTML5 engine and stripped by rehype-sanitize.
When the LLM writes JSX fragments like <Component prop={val} />
outside code fences, remark-parse tags them as html nodes.
rehype-raw then feeds them to parse5, and rehype-sanitize strips
the unknown elements, silently destroying content. Without
rehype-raw, Streamdown auto-injects a remark plugin that converts
html nodes to text, preserving them as visible escaped text.
Markdown formatting (bold, italic, links, code blocks, tables)
is unaffected since those go through remark/rehype directly.
## Summary
- **Removed** `docs/install/cloud/ec2.md` — the standalone EC2 install
guide.
- **Renamed** `docs/install/cloud/aws-mktplc-ce.md` →
`docs/install/cloud/aws-marketplace.md` for a clearer, more discoverable
filename.
- **Updated** `docs/manifest.json`: replaced the "AWS EC2" entry with
"AWS Marketplace" pointing to the renamed file.
- **Updated** `docs/install/cloud/index.md`: fixed the internal link to
the renamed file.
* Adds a "molly-guard" to require users to type the workspace name
before the 'Archive & delete workspace' action fires. This prevents
accidental deletion of 'pet' workspaces.
* This is only shown for workspaces created *before* the chat was
created. The logic here is that any workspace that existed previous to
the chat *cannot* have been created by the chat.
Eliminates the timing flake in
`TestInterruptAutoPromotionIgnoresLaterUsageLimitIncrease` by making the
chatd worker loop clock-controllable.
## Changes
**`coderd/chatd/chatd.go`**
- Replace `time.NewTicker` calls in `Server.start()` with
`p.clock.NewTicker` using named quartz tags `("chatd", "acquire")` and
`("chatd", "stale-recovery")`.
**`coderd/chatd/chatd_test.go`**
- Inject `quartz.NewMock(t)` into the test via `newActiveTestServer`
config override.
- Trap the acquire ticker so the test controls exactly when pending
chats are reacquired.
- Rewrite the test flow as explicit clock-advance steps instead of
wall-clock polling.
**`AGENTS.md`**
- Document the PR title scope rule (scope must be a real path containing
all changed files).
## Validation
- `go test ./coderd/chatd -run
TestInterruptAutoPromotionIgnoresLaterUsageLimitIncrease -count=100` ✅
- `go test ./coderd/chatd` ✅
- `make lint` ✅
## Problem
The VS Code extension embeds the Coder agent chat UI in an iframe,
passing the session token via `postMessage`. HTTP requests use the
`Coder-Session-Token` header, but browser WebSocket connections **cannot
carry custom headers** — they rely on cookies. This causes all WebSocket
requests (e.g. streaming chat messages) to fail with authorization
errors in the embedded iframe.
## Solution
Add `POST /api/v2/users/me/session/token-to-cookie` — a lightweight
endpoint that converts the current (already-validated) session token
into a `Set-Cookie` response. The frontend embed bootstrap flow calls
this immediately after `API.setSessionToken(token)`, before any
WebSocket connections are opened.
### Backend (`coderd/userauth.go`, `coderd/coderd.go`)
- New handler `postSessionTokenCookie` behind `apiKeyMiddleware`.
- Reads the validated token via `httpmw.APITokenFromRequest(r)`.
- Sets an `HttpOnly` cookie with the API key's expiry, applying
site-wide cookie config (Secure, SameSite, host prefix) via
`HTTPCookies.Apply`.
- Returns `204 No Content`.
### Frontend (`site/src/pages/AgentsPage/EmbedContext.tsx`)
- `bootstrapChatEmbedSessionFn` now calls the new endpoint after setting
the header token and before fetching user/permissions.
- The cookie is in place before any WebSocket connections are opened.
## Security
- **No privilege escalation**: The token is already valid — this just
moves it from a header credential to a cookie credential.
- **POST only**: Avoids CSRF-via-navigation.
- **Same origin**: The iframe loads from the Coder server, so the cookie
applies to the correct domain.
- **HttpOnly**: The cookie is not accessible to JavaScript.
> Built with [Coder Agents](https://coder.com/agents) 🤖
LazyFileDiff memo comparator: ignore renderAnnotation reference
changes when the file has no lineAnnotations, preventing comment-box
interactions from re-rendering every file diff.
useGitWatcher stale socket guard: check socketRef.current against
the local socket variable in all event handlers. Stale close events
from superseded connections no longer clobber the active socket or
schedule spurious reconnects.
useGitWatcher field comparison guard: add a compile-time Record
type that errors when WorkspaceAgentRepoChanges gains a field not
covered by the bailout comparison.
Tests: stale close race, reference stability on duplicate messages,
per-field change detection (branch, remote_origin, unified_diff),
and no-op removal of unknown repos.
Follow-up to #23243
Three changes that eliminate unnecessary re-renders cascading into
the FileDiff Shadow DOM components during chat/git-watcher updates:
useGitWatcher: compare repo fields before updating state, return
prev Map when nothing changed instead of always allocating a new one.
RemoteDiffPanel: remove dataUpdatedAt from parsedFiles memo deps,
replace it with a content-derived version counter. The memo now only
recomputes when the actual diff string changes.
DiffViewer: pre-compute per-file options and line annotations into
memoized Maps, wrap LazyFileDiff in React.memo so it skips renders
when props are reference-equal.
## Summary
Adds the database schema, API endpoints, SDK types, and encryption
wrappers for admin-managed MCP (Model Context Protocol) server
configurations that chatd can consume. This is the backend foundation
for allowing external MCP tools (Sentry, Linear, GitHub, etc.) to be
used during AI chat sessions.
## Database
Two new tables:
- **`mcp_server_configs`**: Admin-managed server definitions with URL,
transport (Streamable HTTP / SSE), auth config (none / OAuth2 / API key
/ custom headers), tool allow/deny lists, and an availability policy
(`force_on` / `default_on` / `default_off`). Includes CHECK constraints
on transport, auth_type, and availability values.
- **`mcp_server_user_tokens`**: Per-user OAuth2 tokens for servers
requiring individual authentication. Cascades on user/config deletion.
New column on `chats` table:
- **`mcp_server_ids UUID[]`**: Per-chat MCP server selection, following
the same pattern as `model_config_id` — passed at chat creation,
changeable per-message with nil-means-no-change semantics.
## API Endpoints
All routes are under `/api/experimental/mcp/servers/` and gated behind
the `agents` experiment.
**Admin endpoints** (`ResourceDeploymentConfig` auth):
- `POST /` — Create MCP server config
- `PATCH /{id}` — Update MCP server config (full-replace)
- `DELETE /{id}` — Delete MCP server config
**Authenticated endpoints** (all users, enabled servers only for
non-admins):
- `GET /` — List configs (admins see all, members see enabled-only with
admin fields redacted)
- `GET /{id}` — Get config by ID (with `auth_connected` populated
per-user)
**OAuth2 per-user auth flow:**
- `GET /{id}/oauth2/connect` — Initiate OAuth2 flow (state cookie CSRF
protection)
- `GET /{id}/oauth2/callback` — Handle OAuth2 callback, store tokens
- `DELETE /{id}/oauth2/disconnect` — Remove stored OAuth2 tokens
## Security
- **Secrets never returned**: `OAuth2ClientSecret`, `APIKeyValue`, and
`CustomHeaders` are never in API responses — only boolean indicators
(`has_oauth2_secret`, `has_api_key`, `has_custom_headers`).
- **Field redaction for non-admins**: `convertMCPServerConfigRedacted`
strips `OAuth2ClientID`, auth URLs, scopes, and `APIKeyHeader` from
non-admin responses.
- **dbcrypt encryption at rest**: All 5 secret fields use `dbcrypt_keys`
encryption with full encrypt-on-write / decrypt-on-read wrappers (11
dbcrypt method overrides + 2 helpers), following the same pattern as
`chat_providers.api_key`.
- **OAuth2 CSRF protection**: State parameter stored in `HttpOnly`
cookie with `HTTPCookies.Apply()` for correct `Secure`/`SameSite` behind
TLS-terminating proxies.
- **dbauthz authorization**: All 18 querier methods have authorization
wrappers. Read operations use `ActionRead`, write operations use
`ActionUpdate` on `ResourceDeploymentConfig`.
## Governance Model
| Control | Implementation |
|---------|---------------|
| **Global kill switch** | `enabled` defaults to `false` |
| **Availability policy** | `force_on` (always injected), `default_on`
(pre-selected), `default_off` (opt-in) |
| **Per-chat selection** | `mcp_server_ids` on `CreateChatRequest` /
`CreateChatMessageRequest` |
| **Auth gate** | OAuth2 servers require per-user auth before tools are
injected |
| **Tool-level allow/deny** | Arrays on `mcp_server_configs` for
granular tool filtering |
| **Secrets encrypted at rest** | Uses `dbcrypt_keys` (same pattern as
`chat_providers.api_key`) |
## Tests
8 test functions covering:
- Full CRUD lifecycle (create, list, update, delete)
- Non-admin visibility filtering (enabled-only, field redaction)
- `auth_connected` population for OAuth2 vs non-OAuth2 servers
- Availability policy validation (valid values + invalid rejection)
- Unique slug enforcement (409 Conflict)
- OAuth2 disconnect idempotency
- Chat creation with `mcp_server_ids` persistence
## Known Limitations (Deferred)
These are documented and intentional for an experimental feature:
- **Audit logging** not yet wired — will add when feature stabilizes
- **Cross-field validation** (e.g., OAuth2 fields required when
`auth_type=oauth2`) — admin-only endpoint, will add when stabilizing
- **`force_on` auto-injection** — query exists but not yet wired into
chatd tool injection (follow-up)
- **Additional test coverage** — 403 auth tests, GET-by-ID tests,
callback CSRF tests planned for follow-up
## What's NOT in this PR
- Frontend UI (admin panel + chat picker)
- Actual MCP client connections (`chatd/chatmcp/` manager)
- Tool injection into `chatloop/`
Always deploy from main for now. We want to keep testing commits to main
as soon as they're merged, so we're going to disable the release
freezing behavior. We will test cut releases on a separate deployment,
upgraded manually.
Two issues caused the promoted message to never appear:
1. handlePromoteQueuedMessage discarded the ChatMessage returned by
the promote API, relying on the WebSocket to deliver it.
2. Even when the WebSocket did deliver it (via upsertDurableMessage),
the queue_update event in the same batch called
updateChatQueuedMessages, which mutated the React Query cache. This
gave chatMessagesList a new reference, triggering the message sync
effect. The effect found the promoted message in the store but not in
the REST-fetched data, classified it as a stale entry (the path
designed for edit truncation), and called replaceMessages, wiping it.
Fix (1): capture the ChatMessage from the promote response and upsert
it into the store, matching handleSend for non-queued messages.
Fix (2): track the fetched message array elements across effect runs
using element-level reference comparison. Only run the
hasStaleEntries/replaceMessages path when the message objects actually
changed (e.g. a refetch producing new objects from the server), not
when only an unrelated field like queued_messages caused the query
data reference to update. Element references work because
useMemo(flatMap) preserves object identity when only non-message
fields change in the page data.
Updates the `charm.land/fantasy` replace to the rebased `cj/go1.25`
branch on `kylecarbs/fantasy`, which now includes:
- **chore: downgrade to Go 1.25**
- **feat: anthropic computer use**
- **chore: use kylecarbs/openai-go fork for coder/coder compat**
Switches the `openai-go/v3` replace from `SasSwart/openai-go` →
`kylecarbs/openai-go`, which is the same SasSwart perf fork plus a fix
for `WithJSONSet` being clobbered by deferred body serialization.
Without the fix, `NewStreaming` silently drops `stream: true` from
requests. See https://github.com/kylecarbs/openai-go/pull/2 for details.
Updates the editing banner message in the agents chat input from:
> Editing message — all subsequent messages will be deleted
to:
> Editing will delete all subsequent messages and restart the
conversation here.
---
PR generated with Coder Agents
- Add `RequireExperimentWithDevBypass` middleware to
`/.well-known/oauth-authorization-server` and
`/.well-known/oauth-protected-resource` routes, matching the existing
`/oauth2` routes.
- Clients can now detect OAuth2 support via unauthenticated discovery
(404 = not available).
Fixes#21608
## Summary
- guard Agent pages against malformed model provider/model values before
trimming
- reuse a shared model-ref normalizer across Agent detail, sidebar,
list, and create flows
- add regression coverage for malformed catalog and config entries
## Validation
- `cd site && pnpm exec vitest run
src/pages/AgentsPage/modelOptions.test.ts
src/pages/AgentsPage/AgentDetail.test.ts`
- `cd site && pnpm lint:types`
## Problem
Scaletest follow-up storms showed that the chat stream path was doing a
same-replica DB reread for every durable message it had already
delivered locally.
In a 600-chat / 10-turn run, `/stream`-attributed
`GetChatMessagesByChatID` calls reached about 14.2k across 5,400
follow-up turns — roughly **2.63 rereads per turn**. The primary coderd
replicas saturated their DB pools at 60/60 open connections during the
storm window.
The root cause: when pubsub was active, `Subscribe()` suppressed local
durable `message` events and relied entirely on pubsub notify →
`GetChatMessagesByChatID` for catch-up. Same-replica subscribers paid
the full DB round-trip even though the persisting process was on the
same replica.
## Solution
Add a bounded per-chat **durable message cache** to `chatStreamState` so
that same-replica subscribers can catch up from memory instead of the
database.
### How it works
1. `publishMessage()` caches the SDK event in `chatStreamState` before
local fanout and pubsub notify.
2. `publishEditedMessage()` replaces the cache with only the edited
message, then publishes `FullRefresh`.
3. `Subscribe()` handles ordinary `AfterMessageID` notifies by first
consulting the per-chat durable cache and only falling back to
`GetChatMessagesByChatID` on cache miss.
4. `FullRefresh` always forces a DB reread (cache is bypassed).
### Safety properties
- If the cache misses (e.g. message expired or remote replica), the DB
catch-up still runs — no silent message loss.
- `FullRefresh` (edits) always rereads from the database.
- Remote replicas still use the pubsub + DB path unchanged.
- The cache is bounded (`maxDurableMessageCacheSize = 256`) and scoped
per chat — no unbounded memory growth.
## Impact
This change removes the entire same-replica portion of the stream
rereads. Based on the 600-chat follow-up run, the upper bound on saved
work is the same-replica share of about 14.2k `GetChatMessagesByChatID`
rereads, with the observed total stream reread rate at about 2.63
rereads per follow-up turn.
React's [Why You Might Not Need An
Effect](https://react.dev/learn/you-might-not-need-an-effect) article
describes several antipatterns and footguns you might encounter when
working with useEffect, so I fed it to an agent and let it determine
some low hanging fruit to fix:
Replace useEffect+setState patterns with direct computations where the
values are purely derived from props, state, or query data:
- useWebpushNotifications: derive `enabled` inline from query data
instead of setting it via useEffect
- ProxyContext: replace `proxy` state + updateProxy callback + useEffect
with a single useMemo over its three inputs
- useSyncFormParameters: replace ref-sync useEffect with direct
assignment during render
- AgentsSidebar (LoadMoreSentinel): replace two ref-sync useEffects with
direct assignments during render
Co-authored by Coder Agent 🤖
- Adds a new API endpoint `GET /api/v2/users/oidc-claims` that returns
only the **merged claims** (not the separate id_token/userinfo
breakdown). Scoped exclusively to the authenticated user's own identity
— no user parameter, so users cannot view each other's claims.
- Adds a new CLI command:** `coder users oidc-claims` that hits the
above endpoint.
- The existing owner-only debug endpoint is preserved unchanged for
admins who need the full claim breakdown.
> 🤖 This PR was created with the help of Coder Agents, and will be
reviewed by my human. 🧑💻
Added new AWS install documentation and screenshots to support
deployment of AWS Marketplace Coder Community Edition, as the
primary/recommended method on AWS for POCs and experimenting with Coder.
This PR fixes a bug where if a tool result contained binary data it
wouldn't be persisted to the database.
`jsonb` in Postgres is unable to store null bytes which are sometimes
output by tool results. This change makes it so that we encode them with
a special escape sequence before saving them to the database, and decode
them on read.
<img width="808" height="637" alt="Screenshot 2026-03-11 at 13 14 06"
src="https://github.com/user-attachments/assets/9be353eb-ff26-40ec-9f0a-195022b11f43"
/>
## Problem
The `/agents/settings/insights` page had several issues:
1. **Duplicate PRs** in "Recent Pull Requests" — multiple chats
referencing the same PR URL each produced a row
2. **Wildly wrong costs** — the cost subquery summed ALL messages across
the entire chat *tree* (`GROUP BY root_chat_id`), so every chat in a
tree got the same inflated total. When aggregated, the same tree cost
was counted N× per PR in that tree
3. **UI clutter** — too many stat cards, too many table columns, mixed
naming conventions
## Fix
### Backend (SQL)
- **Deduplicate by PR URL** using `DISTINCT ON (COALESCE(cds.url,
c.id::text))` across all 4 queries
- **Fix cost computation**: use two CTEs — `pr_costs` sums cost from ALL
chats that reference a PR (so review chats contribute), `deduped` picks
one row per PR for state/additions/deletions via DISTINCT ON
- **Tests**: 3 subtests covering multi-chat cost summing, different PRs
no duplication, and duplicate URL counted once
### Frontend
- **3 stat cards** (down from 5): Merged, Merge rate, Cost / merge
- **2-line chart** (down from 3): created (dashed) + merged (solid)
- **4-column model table** (down from 7): Model, Merged, Merge rate,
Cost/merge
- **4-column recent table** (down from 7): Title, Status, Cost, Created
— with `table-fixed` to prevent overflow
- **Consistent naming**: no mixed PR/PRs abbreviation, contextual labels
since page title establishes context
- coderd: Wires `options.WorkspaceUsageTracker` into the chatd config.
- chatd: Adds `UsageTracker` and calls `UsageTracker.Add(workspaceID)`
on each heartbeat tick
- chatd: adds tests to verify `last_used_at` bump behaviour
> 🤖 This PR was created with the help of Coder Agents, and will be
reviewed by my human. 🧑💻
ChatMessagePart uses a flat struct with omitempty on all fields,
but some fields are required in their TypeScript variant (no ?
suffix in the variants struct tag). When Go omits a zero-valued
required field, the frontend receives undefined where it expects
a concrete value.
Remove omitempty from fields that are required in at least one
variant: Text, URL, MediaType, FileName, StartLine, EndLine,
Content. Fields where all variants use ? keep omitempty.
Add a sub-test to TestChatMessagePartVariantTags that enforces
this invariant via reflection so future additions cannot
reintroduce the mismatch.
Supersedes #23249
Adds a `deleted` boolean column to the `chat_messages` table. Messages
are never physically deleted from the database — instead they are marked
as deleted so that usage and cost data is preserved.
## Changes
### Migration
- New migration (000444) adds `deleted boolean NOT NULL DEFAULT false`
to `chat_messages`
### SQL queries
- `DeleteChatMessagesAfterID` → `SoftDeleteChatMessagesAfterID` (UPDATE
SET deleted=true instead of DELETE)
- New `SoftDeleteChatMessageByID` query for single-message soft-delete
- All read queries now filter `deleted = false`:
- `GetChatMessageByID`
- `GetChatMessagesByChatID`
- `GetChatMessagesByChatIDDescPaginated`
- `GetChatMessagesForPromptByChatID` (both CTE and main query)
- `GetLastChatMessageByRole`
- Cost/usage queries (`GetChatCostSummary`, `GetChatCostPerModel`, etc.)
intentionally still include deleted messages to preserve accurate spend
tracking
### EditMessage behavior
- Previously: updated the message content in-place + hard-deleted
subsequent messages
- Now: soft-deletes the original message + soft-deletes subsequent
messages + inserts a new message with the updated content
- This preserves the original message data (tokens, cost, content) in
the database
- Adds a wrapper script at `/usr/local/bin/gh` in the dogfood image that
ensures the GitHub CLI stays authenticated even when tokens expire
during long-running workspace sessions.
Requested by @johnstcn, based on suggestion from @kylecarbs.
Co-authored-by: blink-so[bot] <211532188+blink-so[bot]@users.noreply.github.com>
## Problem
Uploading a file on the `/agents` chat page fails with:
```
Failed to execute 'setRequestHeader' on 'XMLHttpRequest': String contains non ISO-8859-1 code point.
```
This happens when the image filename contains non-ASCII characters (e.g.
CJK characters from macOS screenshots like `スクリーンショット.png`, accented
characters, emoji, etc.). HTTP headers only support ISO-8859-1 code
points, and the filename was being interpolated directly into the
`Content-Disposition` header.
## Fix
Use [RFC 5987](https://datatracker.ietf.org/doc/html/rfc5987)
`filename*=UTF-8''` encoding so the percent-encoded name is always valid
in the header. A static ASCII `filename="file"` fallback is included for
older clients.
The server already uses Go's `mime.ParseMediaType` which decodes
`filename*` automatically, so no backend changes are needed.
### Before
```ts
"Content-Disposition": `attachment; filename="${file.name}"`
```
### After
```ts
"Content-Disposition": `attachment; filename="file"; filename*=UTF-8''${encodeURIComponent(file.name)}`
```
## Testing
Added a server-side test (`TestGetChatFile/UnicodeFilename`) that
uploads with a Japanese filename and verifies it round-trips correctly
through the `Content-Disposition` header.
Go serializes ChatMessagePart.Text with omitempty, so empty
reasoning text (from reasoning_start with no delta) is omitted
from JSON. The frontend receives {type: "reasoning"} with text
as undefined, crashing on .trim() calls.
Mark Text as optional in the reasoning variant via the variants
struct tag. This generates ChatReasoningPart with text?: string
and the frontend falls back to "" via nullish coalescing.
Closes#23245
Fixes#21946
When a startup script fails (exits with non-zero code), the UI displayed
a misleading "Workspace agents are not connected" error even though the
agent is actually connected and functional (SSH works, web terminal
works).
- Extracts the `WorkspaceAlert` component from `Workspace.tsx` to its
own component
- Updates the `WorkspaceAlert` component in `Workspace.tsx` distinguish
correctly between agent disconnection, timeout, shutdown, and startup
script failures.
- Fixes double period bug in the alert description ("the agent has not
connected yet.."
Created on behalf of @kylecarbs
---------
Co-authored-by: blink-so[bot] <211532188+blink-so[bot]@users.noreply.github.com>
Co-authored-by: Cian Johnston <cian@coder.com>
*Disclaimer: implemented by a Coder Agent using Claude Opus 4.6*
Ref:
https://github.com/coder/coder/actions/runs/23251163568/job/67597250364
The deploy workflow Flux reconciliation step failed with no visibility
into what went wrong. Two changes:
- Add `--verbose` to every `flux reconcile` invocation to print
generated objects on failure
- Increase `--timeout` for the four `helmrelease` reconciliations from
the default 5m to 10m
Follow-up to #23220, addressing Cian's review comments:
- **SQL casing**: Uppercase `UNNEST` to match `NULLIF`/`COALESCE`
convention in the query.
- **Builder pattern**: `chatMessage` struct now uses unexported fields
with a `newChatMessage` constructor for required fields (role, content,
visibility, modelConfigID, contentVersion) and chainable builder methods
(`withCreatedBy`, `withCompressed`, `withUsage`, `withContextLimit`,
`withTotalCostMicros`, `withRuntimeMs`) for optional/nullable fields.
- **Batch test in chats_test**: Replaced the `for i := 0; i < 2` loop
with a single batch insert of 2 messages to actually exercise the batch
logic.
- **Multi-message querier test**: Added `BatchInsertMultipleMessages`
test verifying 3-message batch insert with role ordering, sequential
IDs, nullable field semantics (NULL for zero UUIDs and zero ints), and
token/cost assertions.
---------
Co-authored-by: Cian Johnston <cian@coder.com>
## Problem
The `/agents/settings/insights` page was broken because
`InsightsContent` was calling `/api/v2/chats/insights/pull-requests`,
but the backend route is registered under
`/api/experimental/chats/insights/pull-requests` (the entire `/chats`
route block lives under `r.Route("/api/experimental", ...)` in
`coderd.go`).
Every other chat endpoint in the frontend correctly uses
`/api/experimental/chats/...`, but this one was missed.
## Fix
- Added `getPRInsights` method to the API client (`api.ts`) pointing to
`/api/experimental/chats/insights/pull-requests`
- Added a `prInsights` react-query helper in `api/queries/chats.ts`
(matching the pattern of `chatCostUsers`, etc.)
- Updated `InsightsContent.tsx` to use the query helper instead of a raw
`fetch()` with the wrong URL
Processes started via the agent process API inherited the agent's
own working directory (/tmp/coder.xxx) when no WorkDir was
specified. SSH sessions already use a fallback chain: configured
agent directory > $HOME. This wires the same manifest directory
closure into the process manager so the priority is now:
explicit req.WorkDir > agent configured dir > $HOME
The resolved directory is recorded on the process struct so
ProcessInfo.WorkDir and pathStore notifications reflect where
the process actually ran.
Replaces the singular `InsertChatMessage` query with
`InsertChatMessages` that uses PostgreSQL's `unnest()` for batch
inserts. This reduces the number of database round-trips when inserting
multiple messages in a single transaction.
## Changes
- **SQL**: New `InsertChatMessages :many` query using `unnest()` arrays
following the existing codebase pattern (e.g.,
`InsertWorkspaceAgentStats`). Preserves the CTE that updates
`chats.last_model_config_id` using the last non-null model config from
the batch. Uses `NULLIF` for UUID columns to handle NULL foreign keys.
- **Go layers**: Updated `querier.go`, `dbauthz.go`,
`dbmetrics/querymetrics.go`, `dbmock/dbmock.go`, and `queries.sql.go` to
use the new batch signature (`[]ChatMessage` return type, array params).
- **chatd.go**: All call sites converted to batch inserts:
- **CreateChat**: System prompt + user message batched into one call
- **persistStep**: Assistant message + tool messages batched into one
call
- **persistSummary**: Hidden summary + assistant + tool messages batched
into one call
- Single-message sites use the same API with single-element arrays
- **Helper**: New `appendChatMessage` function simplifies building batch
params at each call site.
- **Tests**: All test files updated to use the new API.
Builds on top of #23213.
Two migrations were merged with the same number 000444:
- `000444_usage_events_ai_seats` (#22689, merged first at 09:30) — keeps
000444
- `000444_chat_message_runtime_ms` (#23219, merged second at 10:57) —
renumbered to **000445**
This collision causes `golang-migrate` to fail at runtime since it reads
both files as the same version.
**Fix:** Rename `000444_chat_message_runtime_ms.{up,down}.sql` →
`000445_chat_message_runtime_ms.{up,down}.sql`.
Closes https://github.com/coder/internal/issues/1411
## What
Adds a new admin-only **PR Insights** page for the `/agents` analytics
view — a dashboard for engineering leaders to understand code shipped by
AI agents.
### Backend
- `GET /api/v2/chats/insights/pull-requests` — admin-only endpoint
- 4 SQL queries in `chatinsights.sql` aggregating `chat_diff_statuses`
joined with chat cost data (via root chat tree rollup)
- Runs 5 parallel DB queries: current summary, previous summary (for
trends), time series, per-model breakdown, recent PRs
- SDK types auto-generate to TypeScript
### Frontend (`PRInsightsView`)
- **Stat cards**: PRs created, Merged, Merge rate, Lines shipped,
Cost/merged PR — with trend badges comparing to previous period
- **Activity chart**: Stacked area chart (created/merged/closed) using
git color tokens (`git-added-bright`, `git-merged-bright`,
`git-deleted-bright`)
- **Model performance table**: Per-model PR counts, inline merge rate
bars, diff stats, cost breakdown
- **Recent PRs table**: Status badges, review state icons, author info,
external links
- **Time range filter**: 7d/14d/30d/90d button group
- **4 Storybook stories**: Default, HighPerformance, LowVolume, NoPRs
### Data source
All PR data comes from the existing `chat_diff_statuses` table
(populated by the `gitsync.Worker` background job that polls GitHub
every 120s). No new data collection required.
### Screenshot
View in Storybook: `pages/AgentsPage/PRInsightsView`
## Summary
Adds a `runtime_ms` column to `chat_messages` that records the
wall-clock duration (in milliseconds) of each LLM step. This covers LLM
streaming, tool execution, and retries — the full time the agent is
"alive" for a step.
This is the foundation for billing by agent alive time. The column
follows the same pattern as `total_cost_micros`: stored per assistant
message, aggregatable with `SUM()` over time periods by user.
## Changes
- **Migration**: adds nullable `runtime_ms bigint` to `chat_messages`.
- **chatloop**: adds `Runtime time.Duration` field to `PersistedStep`,
measures `time.Since(stepStart)` at the beginning of each step (covering
stream + tool execution + retries).
- **chatd**: passes `step.Runtime.Milliseconds()` to the assistant
message `InsertChatMessage` call; all other message types (system, user,
tool) get `NULL`.
- **Tests**: adds `runtime > 0` assertion in chatloop tests.
## Billing query pattern
Once ready, aggregation mirrors the existing cost queries:
```sql
SELECT COALESCE(SUM(cm.runtime_ms), 0)::bigint AS total_runtime_ms
FROM chat_messages cm
JOIN chats c ON c.id = cm.chat_id
WHERE c.owner_id = @user_id
AND cm.created_at >= @start_time
AND cm.created_at < @end_time
AND cm.runtime_ms IS NOT NULL;
```
## Description
Implements the server-side merge logic for the `merge_strategy`
attribute added to `coder_env` in [terraform-provider-coder
v2.15.0](https://github.com/coder/terraform-provider-coder/pull/489).
This allows template authors to control how duplicate environment
variable names are combined across multiple `coder_env` resources.
Relates to https://github.com/coder/coder/issues/21885
## Supported strategies
| Strategy | Behavior |
|----------|----------|
| `replace` (default) | Last value wins — backward compatible |
| `append` | Joins values with `:` separator (e.g. PATH additions) |
| `prepend` | Prepends value with `:` separator |
| `error` | Fails the build if the variable is already defined |
## Example
```hcl
resource "coder_env" "path_tools" {
agent_id = coder_agent.dev.id
name = "PATH"
value = "/home/coder/tools/bin"
merge_strategy = "append"
}
```
## Changes
- **Proto**: Added `merge_strategy` field to `Env` message in
`provisioner.proto`
- **State reader**: Updated `agentEnvAttributes` struct and proto
construction in `resources.go`
- **Merge logic**: Added `mergeExtraEnvs()` function in
`provisionerdserver.go` with strategy-aware merging for both agent envs
and devcontainer subagent envs
- **Tests**: 15 unit tests covering all strategies, edge cases (empty
values, mixed strategies, multiple appends)
- **Dependency**: Bumped `terraform-provider-coder` v2.14.0 → v2.15.0
- **Fixtures**: Updated `duplicate-env-keys` test fixtures and golden
files
## Ordering
When multiple resources `append` or `prepend` to the same key, they are
processed in alphabetical order by Terraform resource address (per the
determinism fix in #22706).
When a chat is created via `chatd`, a system message is now inserted
informing the model whether the chat was created with or without a
workspace.
**With workspace:**
> This chat is attached to a workspace. You can use workspace tools like
execute, read_file, write_file, etc.
**Without workspace:**
> There is no workspace associated with this chat yet. Create one using
the create_workspace tool before using workspace tools like execute,
read_file, write_file, etc.
This is a model-only visibility system message (not shown to users) that
helps the model understand its available capabilities upfront —
particularly important for subagents spawned without a workspace, which
previously would attempt to use workspace tools and fail.
**Changes:**
- `coderd/chatd/chatd.go`: Added workspace awareness constants and
inserted the system message in `CreateChat` after the system prompt,
before the initial user message.
- `coderd/chatd/chatd_test.go`: Added
`TestCreateChatInsertsWorkspaceAwarenessMessage` with sub-tests for both
with-workspace and without-workspace cases.
## Summary
- add a hidden deployment config option for chat acquire batch size
(`CODER_CHAT_ACQUIRE_BATCH_SIZE` / `chat.acquireBatchSize`)
- thread the configured value into chatd startup while preserving the
existing default of `10`
- clamp the deployment value to the `int32` range before passing it into
chatd
- regenerate the API/docs/types/testdata artifacts for the new config
field
## Why
`chatd` currently acquires pending chats in batches of `10` via a
compile-time default. This change makes that batch size
operator-configurable from deployment config, so we can tune acquisition
behavior without another code change.
Both message parsers accepted untyped input and relied on scattered
asRecord/asString calls to extract fields at runtime. With the
discriminated ChatMessagePart union, both accept typed input directly
and narrow via switch (part.type).
parseMessageContent narrows from (content: unknown) to
(content: readonly ChatMessagePart[] | undefined), removing legacy
input shape handling the Go backend normalizes away.
applyMessagePartToStreamState narrows from Record<string, unknown>
to ChatMessagePart.
The SSE type guards had a & Record<string, unknown> intersection
that widened everything untyped downstream. Since the data comes
from our own API, the intersection was removed and all handlers in
ChatContext now use generated types directly.
Fixes tool_call_id and tool_name variant tags in codersdk/chats.go:
marked optional to match reality (Go guards against empty values,
omitempty omits them at the wire level).
Refs #23168, #23175
Fixes AIGOV-141
The `coder support bundle` command previously required admin permissions
(`Read DeploymentConfig`) and would abort entirely for non-admin
`member` users with:
```
failed authorization check: cannot Read DeploymentValues
```
This change makes the command **degrade gracefully** instead of failing
outright.
<details>
<summary>
Changes
</summary>
### `support/support.go`
- **`Run()`**: The authorization check for `Read DeploymentValues` is
now a soft warning instead of a hard gate. Unauthenticated users (401)
still fail, but authenticated users with insufficient permissions
proceed with reduced data.
- **`DeploymentInfo()`**: `DeploymentConfig` and `DebugHealth` fetches
now handle 403/401 responses gracefully, matching the existing pattern
used by `DeploymentStats`, `Entitlements`, and `HealthSettings`.
- **`NetworkInfo()`**: Coordinator debug and tailnet debug fetches now
check response status codes for 403/401 before reading the body.
### `cli/support.go`
- **`summarizeBundle()`**: No longer returns early when `Config` or
`HealthReport` is nil. Instead prints warnings and continues summarizing
available data (e.g., netcheck).
### Tests
- `MissingPrivilege` → `MemberNoWorkspace`: Asserts member users can
generate a bundle successfully with degraded admin-only data.
- `NoPrivilege` → `MemberCanGenerateBundle`: Asserts the CLI produces a
valid zip bundle for member users.
- All existing tests continue to pass (`NoAuth`, `OK`, `OK_NoWorkspace`,
`DontPanic`, etc.).
## Behavior matrix
| User type | Before | After |
|---|---|---|
| **Admin** | Full bundle | Full bundle (no change) |
| **Member** | Hard error | Bundle with degraded admin-only data |
| **Unauthenticated** | Hard error | Hard error (no change) |
Related to PRODUCT-182
## Summary
The `AgentsPageView: Opens Analytics For Admins` story was flaky because
the analytics header renders a rolling 30-day date range in the
top-right corner. Since that range was based on the current date, the
story output changed every day.
This change makes the story deterministic by:
- adding an optional `analyticsNow` prop to `AgentsPageView`
- passing that value through to `AnalyticsPageContent` when the
analytics panel is shown
- setting a fixed local-noon timestamp in the story so the rendered
range label stays stable across timezones
## Summary
- include the current agent chat ID in VS Code and Cursor deep links
opened from the agent detail page
- extend `getVSCodeHref` so `chatId` is added only when provided
- add focused tests for deep-link generation with and without `chatId`
## Testing
- `pnpm -C site run format -- src/modules/apps/apps.ts
src/modules/apps/apps.test.ts src/pages/AgentsPage/AgentDetail.tsx`
- `pnpm -C site run check -- src/modules/apps/apps.ts
src/modules/apps/apps.test.ts src/pages/AgentsPage/AgentDetail.tsx`
- `pnpm -C site exec vitest run src/modules/apps/apps.test.ts`
- `pnpm -C site run lint:types`
---
_Generated with [`mux`](https://github.com/coder/mux) • Model:
`openai:gpt-5.4` • Thinking: `high`_
When a PR is detected for a chat, display a compact PR badge in the
AgentDetail TopBar. On mobile it is always visible; on desktop it is
hidden when the sidebar panel is open (which already surfaces PR info)
and shown when the panel is closed.
The badge shows a state-colored icon (open, draft, merged, closed) and
the PR title or number, linking to the PR URL. Only URLs confirmed as
real PRs (via explicit `pull_request_state` or a `/pull/<number>`
pathname) trigger the badge.
## Changes
- **`TopBar.tsx`** — Added `diffStatusData` prop, `PrStateIcon` helper,
and a PR link badge between the title and actions area. Hidden on
desktop when the sidebar panel is open.
- **`AgentDetailView.tsx`** — Pass `diffStatusData` through to
`AgentDetailTopBar`.
- **`TopBar.stories.tsx`** — Added stories for open, draft, merged, and
closed PR states.
Our dogfood image already included chrome. Since we run dogfood
workspaces in Docker, chrome requires some compatibility flags to run
properly. If you launch chrome without them, some webpages crash and
fail to load.
The newest release of https://github.com/coder/portabledesktop added an
icon dock. This PR edits the chrome `.desktop` files so when you open
chrome from the dock it runs with the correct flags.
https://github.com/user-attachments/assets/7bf880e1-22a4-4faa-8f7f-394863c6b127
RenderBlock's file-reference variant diverged from the API (camelCase
vs snake_case), and both file variants were defined inline duplicating
the generated ChatFilePart and ChatFileReferencePart types. The
thinking and file-reference variants carried dead fields (title, text)
that were never populated by the backend.
Replace inline definitions with references to generated types, remove
dead fields, and simplify ReasoningDisclosure (disclosure button path
was dead without title).
Refs #23168
Addresses five documentation gaps identified from an internal agents
briefing Q&A, specifically around what permissions an agent inherits
from the user:
1. **No privilege escalation** — Added explicit statement that the agent
has the exact same permissions as the user. No escalation, no shared
service account.
2. **Cross-user workspace isolation** — Added statement that agents
cannot access workspaces belonging to other users.
3. **Default-state warning** — Added WARNING callouts that agent
workspaces inherit the user's full network access unless templates
explicitly restrict it.
4. **Tool boundary statement** — Added explicit statement that the agent
cannot act outside its defined tool set and has no direct access to the
Coder API.
5. **Template visibility scoped to user RBAC** — Clarified that template
selection respects the user's role and permissions.
Changes across 3 files:
- `docs/ai-coder/agents/index.md`
- `docs/ai-coder/agents/architecture.md`
- `docs/ai-coder/agents/platform-controls/template-optimization.md`
---
PR generated with Coder Agents
## Problem
When a user cancels a streaming chat response mid-stream, the partial
content disappears entirely — both from the UI and the database. The
streamed text vanishes as if the response never happened.
## Root Causes
Three issues combine to prevent partial message persistence on
interrupt:
### 1. StreamPartTypeError only matched `context.Canceled`
(`chatloop.go`)
The interrupt detection in `processStepStream` checked:
```go
errors.Is(part.Error, context.Canceled) && errors.Is(context.Cause(ctx), ErrInterrupted)
```
But some providers propagate `ErrInterrupted` directly as the stream
error rather than wrapping it in `context.Canceled`. This caused the
condition to fail, so `flushActiveState` was never called and partial
text accumulated in `activeTextContent` was lost.
### 2. No post-loop interrupt check (`chatloop.go`)
If the stream iterator stops yielding parts without producing a
`StreamPartTypeError` (e.g., a provider that silently closes the
response body on cancel), there was no check after the `for part :=
range stream` loop to detect the interrupt and flush active state.
### 3. Worker ownership check blocked interrupted persists (`chatd.go`)
`InterruptChat` → `setChatWaiting` clears `worker_id` in the DB
**before** the chatloop detects the interrupt. When
`persistInterruptedStep` (using `context.WithoutCancel`) tried to write
the partial message, the ownership check:
```go
if !lockedChat.WorkerID.Valid || lockedChat.WorkerID.UUID != p.workerID {
return chatloop.ErrInterrupted // always blocks!
}
```
unconditionally rejected the write. The error was silently logged as a
warning.
## Fix
- **Broaden the `StreamPartTypeError` interrupt detection** to match
both `context.Canceled` and `ErrInterrupted` as the stream error.
- **Add a post-loop interrupt check** in `processStepStream` that
flushes active state when the context was canceled with
`ErrInterrupted`.
- **Allow `persistStep` to write when the chat is in `waiting` status**
(interrupt) even if `worker_id` was cleared. The `pending` status (from
`EditMessage`, where history is truncated) still correctly blocks stale
writes.
## Testing
Added `TestInterruptChatPersistsPartialResponse` — an end-to-end
integration test that:
1. Streams partial text chunks from a mock LLM
2. Waits for the chatloop to publish `message_part` events (confirming
chunks were processed)
3. Interrupts the chat mid-stream
4. Verifies the partial assistant message is persisted in the database
with the expected text content
## Summary
- add shared MCP annotation metadata to toolsdk tools
- emit MCP tool annotations from both coderd and CLI MCP servers
- cover annotation serialization in toolsdk, coderd MCP e2e, and CLI MCP
tests
## Why
- Coder already exposed MCP tools, but it did not populate MCP tool
annotation hints (`readOnlyHint`, `destructiveHint`, `idempotentHint`,
`openWorldHint`).
- Hosts such as Claude Desktop use those hints to classify and group
tools, so without them Coder tools can get lumped together.
- This change adds a shared annotation source in `toolsdk` and has both
MCP servers emit those hints through `mcp.Tool.Annotations`, avoiding
drift between local and remote MCP implementations.
## Testing
- Tested locally on Cladue Desktop and the tools are categorized
correctly.
<table>
<tr>
<td> Before
<td> After
<tr>
<td> <img width="613" height="183" alt="image"
src="https://github.com/user-attachments/assets/29d2e3fb-53bc-4ea7-bdb3-f10df4ef996b"
/>
<td> <img width="600" height="457" alt="image"
src="https://github.com/user-attachments/assets/cc384036-c9a7-4db9-9400-43ad51920ff5"
/>
</table>
Note: Done using Coder Agents, reviewed and tested by human locally
The flat ChatMessagePart interface had 20+ optional fields, preventing
TypeScript from narrowing types on switch(part.type). Each consumer
needed runtime validation, type assertions, or defensive ?. chains.
Add `variants` struct tags to ChatMessagePart fields declaring which
union variants include each field. A codegen mutation in apitypings
reads these tags via reflect and generates per-variant sub-interfaces
(ChatTextPart, ChatReasoningPart, etc.) plus a union type alias.
A test validates every field has a variants tag or is explicitly
excluded, and every part type is covered.
Remove dead frontend code: normalizeBlockType, alias case branches
("thinking", "toolcall", "toolresult"), legacy field fallbacks
(line_number, typedBlock.name/id/input/output), and result_delta
handling. Add test coverage for args_delta streaming, provider_executed
skip logic, and source part parsing.
Transient 'No such file or directory' errors from disappearing
overlay2 layers during container operations pollute the displayed
metadata value. Redirect stderr to /dev/null.
Adds a new `site_config` entry that controls whether the virtual desktop
feature for Coder Agents is enabled. It can be set via a new
`/api/experimental/chats/config/desktop-enabled` endpoint, which will be
used by the frontend.
## Summary
Two targeted performance improvements to the chatd server, identified
through benchmarking.
### 1. RWMutex for instruction cache
The instruction cache is read on every chat turn to fetch the home
instruction file for a workspace agent. Writes only occur on cache
misses (once per agent per 5-minute TTL window), making the access
pattern ~90%+ reads.
Switching from `sync.Mutex` to `sync.RWMutex` and using
`RLock`/`RUnlock` on the read path allows concurrent readers instead of
serializing them.
**Benchmark (200 concurrent chats):**
| | ns/op |
|---|---|
| Mutex | 108 |
| RWMutex | 32 |
| **Speedup** | **3.4x** |
### 2. Hoist JSON marshaling out of persistStep transaction
`MarshalParts`, `PartFromContent`, `CalculateTotalCostMicros`, and the
`usageForCost` struct population are pure CPU work that ran inside the
`FOR UPDATE` transaction in `persistStep`. They have zero dependency on
the database transaction.
Moving all marshal and cost-calculation calls above `p.db.InTx()` means
the row lock is held only for `GetChatByIDForUpdate` +
`InsertChatMessage` calls.
**Benchmark (16 goroutines contending on same lock):**
| Tool calls | Inside lock | Outside lock | Speedup |
|---|---|---|---|
| 1 | 13,977 ns/op | 1,055 ns/op | 13x |
| 5 | 38,203 ns/op | 3,769 ns/op | 10x |
| 10 | 67,353 ns/op | 7,284 ns/op | 9x |
| 20 | 145,864 ns/op | 14,045 ns/op | 10x |
No behavioral changes in either commit.
Renames the page title from "Template Routing" to "Template
Optimization" in both the markdown H1 header and the docs manifest
entry.
---
PR generated with Coder Agents
Add a new docs page under /docs/ai-coder/agents/ covering best practices
for creating templates that are discoverable and useful to Coder Agents.
Covers template descriptions, dedicated agent templates, network
boundaries, credential scoping, parameter design, pre-installed tooling,
and prebuilt workspaces for reducing provisioning latency.
<!--
If you have used AI to produce some or all of this PR, please ensure you
have read our [AI Contribution
guidelines](https://coder.com/docs/about/contributing/AI_CONTRIBUTING)
before submitting.
-->
## Summary
Remove the `hidden` tag from the `PromptCacheKey` field on
`ChatModelOpenAIProviderOptions` so the auto-generated JSON schema
no longer marks it as hidden. This allows the admin model
configuration UI to render a "Prompt Cache Key" text input for
OpenAI models alongside other visible options like Reasoning Effort,
Service Tier, and Web Search.
## Changes
- **`codersdk/chats.go`**: Remove `hidden:"true"` from `PromptCacheKey`
struct tag.
- **`site/src/api/chatModelOptionsGenerated.json`**: Regenerated via
`make gen` — `hidden: true` removed from the `prompt_cache_key` entry.
- **`modelConfigFormLogic.test.ts`**: Extend existing "all fields set"
tests to cover extract and build roundtrip for `promptCacheKey`.
## How it works
The `hidden` Go struct tag propagates through the code generation
pipeline:
1. Go struct tag → `scripts/modeloptionsgen` →
`chatModelOptionsGenerated.json`
2. The frontend `getVisibleProviderFields()` filters out fields with
`hidden: true`
3. Removing the tag makes the field visible in the schema-driven form
renderer
No new UI components are needed — the existing `ModelConfigFields`
component
automatically renders the field as a text input based on the schema
(`type: "string"`, `input_type: "input"`).
The field appears as **"Prompt Cache Key"** with description
"Key for enabling cross-request prompt caching" in the OpenAI provider
section of the admin model configuration form.
Fixes#18573
## Changes
When a `coder_app` resource sets `open_in = "tab"`, clicking the app
link now opens in a new browser tab instead of navigating in the same
tab.
`target="_blank"` and `rel="noreferrer"` are set inline on the
`<a>` elements in `AppLink.tsx`, gated on `app.open_in === "tab"`. This
follows the codebase convention of co-locating `target` and `rel` at the
render site.
`noreferrer` suppresses the Referer header to avoid leaking workspace IDs
to destination servers and implies `noopener`.
`noopener` prevents tabnabbing — without it, the opened page can
redirect the Coder dashboard tab via `window.opener`. This is especially
relevant for same-origin path-based apps, which would otherwise have
full DOM access to the dashboard.
> **Future enhancement**: template admins could opt into sending the
referrer via a `coder_app` setting, enabling feedback pages built around
workspace context.
## Tests
A vitest case is added in `AppLink.test.tsx` (rather than a Storybook
story, since the assertions are purely behavioral with no visual
component):
- **`sets target=_blank and rel=noopener noreferrer when open_in is
tab`** — renders the app link with `open_in: "tab"` and asserts
`target="_blank"` and `rel="noreferrer"` are present on the
anchor.
## Slim-window behavior
The `slim-window` test case and the `openAppInNewWindow()` comment in
`apps.ts` have been split out into a follow-up PR for separate review,
since the `window.open()` / `noopener` tradeoffs there deserve dedicated
discussion.
---------
Co-authored-by: Kayla はな <kayla@tree.camp>
Introduce a three-way workspace sharing setting (none, everyone,
service_accounts) replacing the boolean workspace_sharing_disabled.
In service_accounts mode, only service account-owned workspaces can be
shared while regular members' share permissions are removed. Adds a
new organization-service-account system role with per-org permissions
reconciled alongside the existing organization-member system role.
Related to:
https://linear.app/codercom/issue/PLAT-28/feat-service-accounts-sharing-mode-and-rbac-role
---------
Co-authored-by: Steven Masley <Emyrk@users.noreply.github.com>
Co-authored-by: Kayla はな <mckayla@hey.com>
## Summary
Fixes four interrelated issues that caused scroll position jumps and
phantom scroll growth when paginating older chat messages.
## Changes
### 1. Removed client-side message windowing (`useMessageWindow`)
There were two competing sentinel systems: server-side pagination and
client-side windowing. The client windowing sentinel was nested deep
inside the timeline with no explicit IntersectionObserver `root`,
causing scroll position jumps when messages were prepended. Blink
(coder/blink) has no client-side windowing. Removed it entirely; server
pagination + `contentVisibility` handled performance.
### 2. Removed `contentVisibility: "auto"` from message sections
Each section had `contentVisibility: "auto"` with `containIntrinsicSize:
"1px 600px"`, causing the scroll region to grow/shrink as the browser
swapped 600px placeholders for actual heights while scrolling. This
created phantom scroll growth with no fetch involved.
### 3. Gated WebSocket on initial REST data
The WebSocket `Subscribe` snapshot calls `GetChatMessagesByChatID` (no
LIMIT) which returns every message when `afterMessageID` is 0. The
WebSocket effect opened before the REST page resolved, so
`lastMessageIdRef` was undefined, causing the server to replay the
entire history and defeating pagination. Added `initialDataLoaded` guard
so the socket waits for the first REST page.
### 4. Manual scroll position restoration
Replaced unreliable CSS scroll anchoring in `flex-col-reverse` with a
`ScrollAnchoredContainer` that snapshots `scrollHeight` before fetch and
restores `scrollTop` via `useLayoutEffect` after render. Disabled
browser scroll anchoring (`overflow-anchor: none`) to prevent conflicts.
File references in user messages now render as inline chips (matching
the chat input style) instead of in a separate bordered section at the
bottom of the message bubble.
This reimplements #23131 which was accidentally reverted during the
merge of #23072 (the spend-limit UI PR resolved a merge conflict by
dropping the inline chip logic).
## Changes
- **FileReferenceNode.tsx**: Export `FileReferenceChip` so it can be
imported for read-only use (no remove button when `onRemove` is
omitted).
- **ConversationTimeline.tsx**: Iterate through `parsed.blocks` in
document order, rendering `response` blocks as text and `file-reference`
blocks as inline `FileReferenceChip` components. Removes the old
separated file-reference section with `border-t` divider.
- **ConversationTimeline.stories.tsx**: Added
`UserMessageWithInlineFileRef` and
`UserMessageWithMultipleInlineFileRefs` stories.
## Problem
The chat listing endpoint (`GetChatsByOwnerID`) was using
`fetchWithPostFilter`, which fetches N rows from the database and then
filters them in Go memory using RBAC checks. This causes a pagination
bug: if the user requests `limit=25` but some rows fail the auth check,
fewer than 25 rows are returned even though more authorized rows exist
in the database. The client may incorrectly assume it has reached the
end of the list.
## Solution
Switch to the same pattern used by `GetWorkspaces`, `GetTemplates`, and
`GetUsers`: `prepareSQLFilter` + `GetAuthorized*` variant. The RBAC
filter is compiled to a SQL WHERE clause and injected into the query
before `ORDER BY`/`LIMIT`, so the database returns exactly the requested
number of authorized rows.
Additionally, `GetChatsByOwnerID` is renamed to `GetChats` with
`OwnerID` as an optional (nullable) filter parameter, matching the
`GetWorkspaces` naming convention.
## Changes
| File | Change |
|------|--------|
| `queries/chats.sql` | Renamed to `GetChats`, `owner_id` now optional
via CASE/NULL, added `-- @authorize_filter` |
| `queries.sql.go` | Renamed constant, params struct (`GetChatsParams`),
and method |
| `querier.go` | Interface method renamed |
| `modelqueries.go` | Added `chatQuerier` interface +
`GetAuthorizedChats` impl |
| `dbauthz/dbauthz.go` | `GetChats` now uses `prepareSQLFilter` instead
of `fetchWithPostFilter` |
| `dbauthz/dbauthz_test.go` | Updated tests for SQL filter pattern |
| `dbmock/dbmock.go` | Renamed + added mock for `GetAuthorizedChats` |
| `dbmetrics/querymetrics.go` | Renamed + added metrics wrapper |
| `rbac/regosql/configs.go` | Added `ChatConverter` (maps `org_owner` to
empty string literal since `chats` has no `organization_id` column) |
| `rbac/authz.go` | Added `ConfigChats()` |
| `chats.go` | Handler uses renamed method with `uuid.NullUUID` |
| `searchquery/search.go` | Updated return type |
| `gitsync/worker.go` | Updated interface and call site |
| Various test files | Updated for renamed types |
Previously main.go used syscall.SysProcAttr{Setpgid: true} and
syscall.Kill, both undefined on Windows. This broke GOOS=windows
cross-compilation.
Add a //go:build !windows constraint to the package since it is
a dev-only tool that requires Unix utilities (bash, make, etc.)
and is not intended to run on Windows.
Refs #23054Fixescoder/internal#1407
## Problem
The WebSocket handler for `diff_status_change` events in
`AgentsPage.tsx` was triggering a burst of redundant HTTP requests on
every event:
1. **`invalidateChatListQueries(queryClient)`** — Full refetch of the
chats list endpoint. Unnecessary because `updateInfiniteChatsCache`
already writes `diff_status` into the sidebar cache optimistically on
every event.
2. **`invalidateQueries({ queryKey: chatKey(id) })`** — Refetch of the
individual chat. Also unnecessary — the SSE event carries `diff_status`
in its payload and the optimistic updater writes it into the `chatKey`
cache directly. Worse, this call was missing `exact: true`, so TanStack
Query's prefix matching cascaded the invalidation to `chatMessagesKey`,
`chatDiffContentsKey`, and every other query under `["chats", id]`.
Since diff status changes fire frequently during active agent work, this
spammed the chats list endpoint and caused redundant refetches of
messages and diff contents on every single event.
## Fix
Strip the handler down to the one invalidation that's actually needed —
`chatDiffContentsKey` (the file-level diff contents aren't in the SSE
payload):
```typescript
if (chatEvent.kind === "diff_status_change") {
void queryClient.invalidateQueries({
queryKey: chatDiffContentsKey(updatedChat.id),
exact: true,
});
}
```
## Why tests didn't catch this
The existing tests in `chats.test.ts` cover query utilities in isolation
(e.g. `invalidateChatListQueries` scoping, mutation invalidation). The
WebSocket event handler lives in the `AgentsPage` component — there was
no test covering what the `diff_status_change` code path actually
invalidates.
Added regression tests verifying that `exact: true` prevents
prefix-match cascade vs the old behavior.
## Summary
This PR removes two redundant chat rereads in `chatd`.
### Archive / unarchive
- `archiveChat` and `unarchiveChat` already come through
`httpmw.ChatParam`, so the handlers already have the `database.Chat`
row.
- Pass that row into `chatd.ArchiveChat` / `chatd.UnarchiveChat` instead
of rereading by ID before publishing the sidebar events.
### End-of-turn cleanup
- `processChat` no longer calls `GetChatByID` after the cleanup
transaction just to refresh the chat snapshot.
- Title generation already persists the generated title and emits its
own `title_change` event.
- To preserve best-effort title freshness for the cleanup path, the
async title-generation goroutine stores the generated title in per-turn
shared state and cleanup overlays it if available before publishing the
`status_change` event and dispatching push notifications.
## Why
- removes one DB read from archive / unarchive requests
- removes one DB read from completed turns, which is the larger hot-path
win
- keeps the existing pubsub/event contract intact instead of broadening
this into a larger event-model redesign
## Notes
- `title_change` remains the authoritative title update for clients
- cleanup does not wait for title generation; it uses the generated
title only when it is already available
- Add `--starter-template` option and properly create starter template
with name and icon
- Add Coder Desktop URLs to listening banner
- Makefile tweak to avoid rebuilding `scripts/develop` every time Go
code changes
## Summary
- reuse workspace agent context within a single `runChat()` turn
- remove duplicate latest-build agent lookups between
`resolveInstructions()` and `getWorkspaceConn()`
- avoid the extra `GetWorkspaceAgentByID` fetch when the selected
`WorkspaceAgent` already has the needed metadata
- add focused internal tests for reuse and refresh-on-dial-failure
## Why
This came out of a 5000-chat / 10-turn scaletest on bravo against a
single workspace.
The run completed successfully, but coderd stayed DB-pool bound, and one
workspace-backed hot path stood out:
- `GetWorkspaceAgentsInLatestBuildByWorkspaceID ≈ 46.7k`
- `GetWorkspaceByID ≈ 48.0k`
- `GetWorkspaceAgentByID ≈ 2.2k`
Within one `runChat()` turn, chatd was rediscovering the same workspace
agent multiple times just to resolve instructions and open the workspace
connection.
## What this changes
This PR introduces a **turn-local** workspace context helper so a single
acquired turn can:
- resolve the selected workspace agent once
- reuse that agent for instruction resolution
- reuse the same `AgentConn` for workspace tools and reload/compaction
This stays turn-local only, so a later turn on another replica still
rebuilds fresh context from the DB.
## Expected impact
This is an incremental improvement, not a full fix.
It should reduce duplicated workspace-agent lookups and shave some DB
pressure from a hot path for workspace-backed chats, while preserving
multi-replica correctness.
## Testing
- `go test ./coderd/chatd/...`
- `golangci-lint run ./coderd/chatd/...`
## Problem
When `generate.sh` is run (e.g. to regenerate fixtures after adding a
new field like `subagent_id`), the `duplicate-env-keys` fixture gets
UUID scrambling.
The `minimize_diff()` function uses a bash associative array keyed by
JSON field name (`deleted["id"]`). The `duplicate-env-keys` fixture has
multiple `coder_env` resources, each with the same key names (`id`,
`agent_id`). Since an associative array can only hold one value per key,
UUIDs get cross-contaminated or left as random terraform-generated
values.
Discovered while working on #23122.
## Fix
Add `duplicate-env-keys` to the `toskip` array in `generate.sh`,
alongside `kubernetes-metadata`. This fixture uses hand-crafted
placeholder UUIDs and should not be regenerated.
Relates to #21885.
Replace the ad-hoc camelCase file block shape ({ mediaType, fileId, data })
with snake_case fields matching ChatMessagePart from the API types.
The RenderBlock file variant now uses media_type/file_id instead of
mediaType/fileId. The parsers in messageParsing.ts and streamState.ts
pass validated ChatMessagePart objects through directly instead of
destructuring and reassembling with renamed fields. This eliminates
the needless API → camelCase → snake_case roundtrip that the edit
flow previously required.
Refs #22735
handleSubmit (triggered via Enter key) didn't check isUploading, so
messages could be sent while an image upload was still in progress.
The send button was correctly disabled via canSend, but the keyboard
shortcut bypassed that guard.
QueuedMessagesList used untyped extraction helpers that fell through to
JSON.stringify for attachment-only messages. Replace them with a single
getQueuedMessageInfo function using typed ChatMessagePart access.
Show an attachment badge (ImageIcon + count) for file parts, and use a
consistent "[Queued message]" placeholder for all no-text situations.
Editing a queued message with file attachments silently dropped all
attachments because handleStartQueueEdit only accepted text. Thread
file blocks from QueuedMessagesList through the edit callback into
handleStartQueueEdit, which now calls setEditingFileBlocks. The
existing useEffect in AgentDetailInput picks these up and populates
the attachment UI. Also clear editingFileBlocks in handleCancelQueueEdit
and handleSendFromInput.
## Background
A 5000-chat scaletest (~50k turns, ~2m45s wall time) completed
successfully,
but the main bottleneck was **DB pool starvation from repeated reads**,
not
individually expensive SQL. The push/webpush path showed a few
especially noisy
reads:
- `GetLastChatMessageByRole` for push body generation
- `GetEnabledChatProviders` + `GetChatModelConfigByID` for push summary
model
resolution
- `GetWebpushSubscriptionsByUserID` for every webpush dispatch
This PR keeps the optimizations that remove those duplicate reads while
leaving
stream behavior unchanged.
## What changes in this PR
### 1. Reuse resolved chat state for push notifications
`maybeSendPushNotification` used to re-read the last assistant message
and
re-resolve the chat model/provider after `runChat` had already done that
work.
Now `runChat` returns the final assistant text plus the already-resolved
model
and provider keys, and the push goroutine uses that state directly.
That removes the extra push-path reads for:
- `GetLastChatMessageByRole`
- the second `resolveChatModel` path
- the provider/model lookups that came with that second resolution
### 2. Cache webpush subscriptions during dispatch
`Dispatch()` previously hit `GetWebpushSubscriptionsByUserID` on every
push. A
small per-user in-memory cache now avoids those repeated reads.
The follow-up fix keeps that optimization correct: `InvalidateUser()`
bumps a
per-user generation so an older in-flight fetch cannot repopulate the
cache with
pre-mutation data after subscribe/unsubscribe.
That preserves the cache win without letting local subscription changes
be
silently overwritten by stale fetch results.
## Why this is safe
- The push change only reuses data already produced during the same chat
run. It
does not change notification semantics; if there is no assistant text to
summarize, the existing fallback body still applies.
- The webpush change keeps the existing TTL and `410 Gone` cleanup
behavior. The
generation guard only prevents stale in-flight fetches from poisoning
the
shared cache after invalidation.
- The final PR does **not** change stream setup, pubsub/relay behavior,
or chat
status snapshot timing.
## Deliberately not included
- No stream-path optimization in `Subscribe`.
- No inline pubsub message payloads.
- No distributed cross-replica webpush cache invalidation.
Frontend for agent chat spend limiting on `/agents`.
## Changes
- add the limits management UI, API hooks, and validation for
deployment, group, and user overrides
- show spend limit status in Agents analytics and usage summaries
- surface limit-related chat errors consistently in the agent detail
experience
- add shared currency and usage-limit messaging helpers plus related
stories/tests
## Problem
Models frequently use shell `&` instead of `run_in_background=true` when
starting long-running processes through `/agents`, causing them to die
shortly after starting. This happens because:
1. **No guidance in tool schema** — The `ExecuteArgs` struct had zero
`description` tags. The model saw `run_in_background: boolean
(optional)` with no explanation of when/why to use it.
2. **Shell `&` is silently broken** — `sh -c "command &"` forks the
process, the shell exits immediately, and the forked child becomes an
orphan not tracked by the process manager.
3. **No process group isolation** — The SSH subsystem sets `Setsid:
true` on spawned processes, but the agent process manager set no
`SysProcAttr` at all. Signals only hit the top-level `sh`, not child
processes.
## Investigation
Compared our implementation against **openai/codex** and **coder/mux**:
| Aspect | codex | mux | coder/coder (before) |
|--------|-------|-----|---------------------|
| Background flag | Yield/resume with `session_id` | `run_in_background`
with rich description | `run_in_background` with **no description** |
| `&` handling | `setsid()` + `killpg()` | `detached: true` +
`killProcessTree()` | **Nothing** — orphaned children escape |
| Process isolation | `setsid()` on every spawn | `set -m; nohup ...
setsid` for background | **No `SysProcAttr` at all** |
| Signal delivery | `killpg(pgid, sig)` — entire group | `kill -15
-\$pid` — negative PID | `proc.cmd.Process.Signal()` — **PID only** |
## Changes
### Fix 1: Add descriptions to `ExecuteArgs` (highest impact)
The model now sees explicit guidance: *"Use for long-running processes
like dev servers, file watchers, or builds. Do NOT use shell & — it will
not work correctly."*
### Fix 2: Update tool description
The top-level execute tool description now reinforces: *"Use
run_in_background=true for long-running processes. Never use shell '&'
for backgrounding."*
### Fix 3: Detect trailing `&` and auto-promote to background
Defense-in-depth: if the model still uses `command &`, we strip the `&`
and promote to `run_in_background=true` automatically. Correctly
distinguishes `&` from `&&`.
### Fix 4: Process group isolation (`Setpgid`)
New platform-specific files (`proc_other.go` / `proc_windows.go`)
following the same pattern as `agentssh/exec_other.go`. Every spawned
process gets its own process group.
### Fix 5: Process group signaling
`signal()` now uses `syscall.Kill(-pid, sig)` on Unix to signal the
entire process group, ensuring child processes from shell pipelines are
also cleaned up.
## Testing
All existing `agent/agentproc` tests pass. Both packages compile
cleanly.
## Problem
The `build` job on `main` has been failing intermittently (and now
consistently) with `no space left on device` on the
`depot-ubuntu-22.04-8` runner. The runner's disk fills up during Docker
image builds or SBOM generation, depending on how close to the limit a
given run lands.
The build was already at the boundary — the Go build cache alone is ~1.3
GB, build artifacts are ~2 GB, and Docker image builds + SBOM scans need
several hundred MB of headroom in `/tmp`. No single commit caused this;
cumulative growth in dependencies and the scheduled `coder-base:latest`
rebuild on Monday morning nudged it past the limit.
## Fix
Three changes to reclaim ~2 GB of disk before Docker runs:
1. **Build all platform archives and packages in the Build step** —
moves arm64/armv7 `.tar.gz` and `.deb` from the Docker step to the Build
step so we can clean caches in between.
2. **Clean up Go caches between Build and Docker** — once binaries are
compiled, the Go build cache and module cache aren't needed. Also
removes `.apk`/`.rpm` packages that are never uploaded.
3. **Set `DOCKER_IMAGE_NO_PREREQUISITES`** — tells make to skip
redundantly building `.deb`/`.rpm`/`.apk`/`.tar.gz` as prerequisites of
Docker image targets. The Makefile already supports this flag for
exactly this purpose.
## Problem
`updateInfiniteChatsCache`, `prependToInfiniteChatsCache`, and
`readInfiniteChatsCache` use `setQueriesData({ queryKey: ["chats"] })`
which prefix-matches **all** queries starting with `"chats"`, including
`["chats", chatId, "messages"]`.
After #23083 converted chat messages to `useInfiniteQuery`, the cached
messages data gained a `.pages` property containing
`ChatMessagesResponse` objects (not `Chat[]` arrays). The `if
(!prev.pages)` guard no longer bailed out, and the updater called
`.map()` on these objects — `TypeError: Z.map is not a function`.
## Fix
Extract the `isChatListQuery` predicate that already existed inline in
`invalidateChatListQueries` and apply it to all four cache helpers. This
scopes them to sidebar queries (`["chats"]` or `["chats",
<filterOpts>]`) and skips per-chat queries (`["chats", <id>, ...]`).
Adds a `--no-wait` flag (CODER_CREATE_NO_WAIT) to the create command,
matching the existing pattern in `coder start`. When set, the `coder
create` command returns immediately after the workspace creation API
call succeeds instead of streaming build logs until completion.
This enables fire-and-forget workspace creation in CI/automation
contexts (e.g., GitHub Actions), where waiting for the build to finish
is unnecessary. Combined with other existing flags, users can create a
workspace with no interactivity, assuming the user is already
authenticated.
When a user uses an AI feature, we record them in the `ai_seat_state` as consuming a seat.
Added in debouching to prevent excessive writes to the db for this feature. There is no need for frequent updates.
Creates a new table `ai_seat_state` to keep track of when users consume an ai_seat. Once a user consumes an AI seat, they will forever in this table (as it stands today).
## VS Code iframe embed auth via postMessage + setSessionToken
Adds embed auth for VS Code iframe integration, allowing the Coder agent
chat UI to be embedded in VS Code webviews without manual login — using
direct header auth instead of cookies.
### How it works
1. **Parent frame** (VS Code webview) loads an iframe pointing to
`/agents/:agentId/embed`
2. **Embed page** detects the user is signed out and posts
`coder:vscode-ready` to the parent
3. **Parent** responds with `coder:vscode-auth-bootstrap` containing the
user's Coder API token
4. **Embed page** calls `API.setSessionToken(token)` to set the
`Coder-Session-Token` header on all subsequent axios requests
5. **Embed page** fetches user + permissions, sets them in the React
Query cache atomically, and renders the authenticated agent chat UI
No cookies, no CSRF, no backend endpoint needed. The token is passed via
postMessage and used as a header on every API request.
### What changed
**Frontend** (`site/src/pages/AgentsPage/`):
- `AgentEmbedPage.tsx` — added postMessage bootstrap directly in the
embed page: listens for `coder:vscode-auth-bootstrap`, calls
`API.setSessionToken(token)`, fetches user/permissions atomically to
avoid race conditions
- `EmbedContext.tsx` — React context signaling embed mode (from previous
commit, unchanged)
- `AgentDetail/TopBar.tsx` — conditionally hides navigation elements in
embed mode (from previous commit, unchanged)
- Both `/agents/:agentId/embed` and `/agents/:agentId/embed/session`
routes live outside `RequireAuth`
**Auth bootstrap** (`site/src/api/queries/users.ts`):
- `bootstrapChatEmbedSessionFn` now calls `API.setSessionToken(token)`
instead of posting to a backend endpoint
- Fetches user and permissions directly via `API.getAuthenticatedUser()`
and `API.checkAuthorization()`, then sets both in the query cache
atomically — this avoids a race where `isSignedIn` flips before
permissions are loaded
**Removed** (no longer needed):
- `coderd/embedauth.go` — the `POST
/api/experimental/chats/embed-session` handler
- `coderd/embedauth_test.go` — backend tests for the endpoint
- `codersdk/embedauth.go` — `EmbedSessionTokenRequest` SDK type
- `site/src/api/api.ts` — `postChatEmbedSession` method
- `docs/user-guides/workspace-access/vscode-embed-auth.md` — doc page
for the old cookie flow
- Swagger/API doc entries for the endpoint
### Why not cookies?
The initial implementation used a backend endpoint to set an HttpOnly
session cookie. This required `SameSite=None; Secure` for cross-origin
iframes, which doesn't work over HTTP in development (Chrome requires
HTTPS for `Secure` cookies). The `setSessionToken` approach bypasses
cookies entirely — the token is set as an axios default header, and
header-based auth also naturally bypasses CSRF protection.
### Dogfooding
Tested end-to-end with a VS Code extension that:
1. Registers a `/openChat` deep link handler
(`vscode://coder.coder-remote/openChat?url=...&token=...&agentId=...`)
2. Starts a local HTTP reverse proxy (to work around VS Code webview
iframe sandboxing)
3. Loads `/agents/:agentId/embed` in an iframe through the proxy
4. Relays the postMessage handshake between the iframe and the extension
host
5. The embed page receives the token, calls `setSessionToken`, and
renders the chat
Verified: chat title, messages, and input field all display correctly in
VS Code's secondary sidebar panel.
Adds cursor-based pagination to the chat messages endpoint.
## Backend
- New `GetChatMessagesByChatIDPaginated` SQL query: returns messages in
`id DESC` order with a `before_id` keyset cursor and configurable
`limit`
- Handler parses `?before_id=N&limit=N` query params, uses the `LIMIT
N+1` trick to set `has_more` without a separate COUNT query
- Queued messages only returned on the first page (no cursor) since
they're always the most recent
- SDK client updated with `ChatMessagesPaginationOptions`
- Fully backward compatible: omitting params returns the 50 newest
messages
## Frontend
- Switches `getChatMessages` from `useQuery` to `useInfiniteQuery` with
cursor chaining via `getNextPageParam`
- Pages flattened and sorted by `id` ascending for chronological display
- `MessagesPaginationSentinel` component uses `IntersectionObserver`
(200px rootMargin prefetch) inside the existing `flex-col-reverse`
scroll container
- `flex-col-reverse` handles scroll anchoring natively when older
messages are prepended — no manual `scrollTop` adjustment needed (same
pattern as coder/blink)
## Why cursor-based instead of offset/limit
Offset-based pagination breaks when new messages arrive while paginating
backward (offsets shift, causing duplicates or missed messages). The
`before_id` cursor is stable regardless of inserts — each page is
deterministic.
## Problem
The `edit_files` tool used `strings.ReplaceAll` for exact substring
matches, silently replacing **every** occurrence. When an LLM's search
string wasn't unique in the file, this caused unintended edits. Fuzzy
matches (passes 2 and 3) only replaced the first occurrence, creating
inconsistent behavior. Zero matches were also silently ignored.
## Investigation
Investigated how **coder/mux** and **openai/codex** handle this:
| Tool | Multiple matches | No match | Flag |
|---|---|---|---|
| **coder/mux** `file_edit_replace_string` | Error (default
`replace_count=1`) | Error | `replace_count` (int, default 1, -1=all) |
| **openai/codex** `apply_patch` | Uses first match after cursor
(structural disambiguation via context lines + `@@` markers) | Error |
None (different paradigm) |
| **coder/coder** `edit_files` (before) | Exact: replaces all. Fuzzy:
replaces first. | Silent success | None |
## Solution
Adopted the mux approach (error on ambiguity) with a simpler
`replace_all: bool` instead of `replace_count: int`:
- **Default (`replace_all: false`)**: search string must match exactly
once. Multiple matches → error with guidance: *"search string matches N
occurrences. Include more surrounding context to make the match unique,
or set replace_all to true"*
- **`replace_all: true`**: replaces all occurrences (opt-in for
intentional bulk operations like variable renames)
- **Zero matches**: now returns an error instead of silently succeeding
Chose `bool` over `int` count because:
1. LLMs are bad at counting occurrences
2. The real intent is binary (one specific spot vs. all occurrences)
3. Simpler error recovery loop for the LLM
## Changes
| File | Change |
|---|---|
| `codersdk/workspacesdk/agentconn.go` | Add `ReplaceAll bool` to
`FileEdit` struct |
| `agent/agentfiles/files.go` | Count matches before replacing; error if
>1 and not opted in; error on zero matches; add `countLineMatches`
helper |
| `codersdk/toolsdk/toolsdk.go` | Expose `replace_all` in tool schema
with description |
| `agent/agentfiles/files_test.go` | Update existing tests, add
`EditEditAmbiguous`, `EditEditReplaceAll`, `NoMatchErrors`,
`AmbiguousExactMatch`, `ReplaceAllExact` |
## Summary
When the email address returned from an OIDC provider doesn't match the
configured allowed domain list (or isn't verified), users previously saw
raw JSON dumped directly in the browser — an ugly and confusing
experience during a browser-redirect flow.
This PR replaces those JSON responses with the same styled static HTML
error page already used for group allow-list errors, signups-disabled,
and wrong-login-type errors.
## Changes
### `coderd/userauth.go`
Replaced 3 `httpapi.Write` calls in `userOIDC` with
`site.RenderStaticErrorPage`:
| Error case | Title shown |
|---|---|
| Email domain not in allowed list | "Unauthorized email" |
| Malformed email (no `@`) with domain restrictions | "Unauthorized
email" |
| `email_verified` is `false` | "Email not verified" |
All render HTTP 403 with `HideStatus: true` and a "Back to login" action
button.
### `coderd/userauth_test.go`
- Updated `AssertResponse` callbacks on existing table-driven tests
(`EmailNotVerified`, `NotInRequiredEmailDomain`,
`EmailDomainForbiddenWithLeadingAt`) to verify HTML Content-Type and
page content.
- Extended `TestOIDCDomainErrorMessage` to additionally assert HTML
rendering.
- Added new `TestOIDCErrorPageRendering` with 3 subtests covering all
error scenarios, verifying: HTML doctype, expected title/description,
"Back to login" link, and absence of JSON markers.
---------
Co-authored-by: Copilot Autofix powered by AI <175728472+Copilot@users.noreply.github.com>
The minimize_diff function in generate.sh preserves autogenerated
values (UUIDs, tokens, etc.) across regeneration to keep diffs
minimal. The subagent_id field was missing from the preservation
list, causing unnecessary churn on devcontainer test data.
Also regenerates all testdata with the current terraform (1.14.5)
and coder provider (2.14.0).
The completion chime on `/agents` was enabled by default for new users
(or when no localStorage preference existed). This changes the default
to disabled, so users must explicitly opt in via the sound toggle
button.
## Changes
- `getChimeEnabled()` now returns `false` when no preference is stored
(was `true`)
- `catch` fallback also returns `false` (was `true`)
- Updated tests to reflect the new default and explicitly enable the
chime in `maybePlayChime` tests
## Problem
The sidebar diff status (PR icon, +additions/-deletions, file count) was
not updating in real-time. Users had to reload the page to see changes.
Two root causes:
1. **Frontend**: The `diff_status_change` WebSocket handler in
`AgentsPage.tsx` had an early `return` (line 398) that skipped
`updateInfiniteChatsCache`, so the sidebar's cache was never updated.
Even for other event types, the cache merge only spread `status` and
`title` — never `diff_status`.
2. **Server**: `publishChatPubsubEvent` in `chatd.go` constructed a
minimal `Chat` payload without `DiffStatus`, so even if the frontend
consumed the event, `updatedChat.diff_status` would be `undefined`.
## Fix
### Server (`coderd/chatd/chatd.go`)
- `publishChatPubsubEvent` now accepts an optional
`*codersdk.ChatDiffStatus` parameter; when non-nil it's set on the
outgoing `Chat` payload.
- `PublishDiffStatusChange` fetches the diff status from the DB,
converts it, and passes it through.
- Added `convertDBChatDiffStatus` (mirrors `coderd/chats.go`'s converter
to avoid circular import).
- All other callers pass `nil`.
### Frontend (`site/src/pages/AgentsPage/AgentsPage.tsx`)
- Removed the early `return` so `diff_status_change` events fall through
to the cache update logic.
- Added `isDiffStatusEvent` flag and spread `diff_status` into both the
infinite chats cache (sidebar) and the individual chat cache.
The `/archive` and `/desktop` chat endpoints had swagger route comments
(`@Summary`, `@ID`, `@Router`, etc.) that would cause them to appear in
generated API docs. Since these live under `/experimental/chats`, they
should not be documented.
This removes the swagger annotations and adds the standard `//
EXPERIMENTAL: this endpoint is experimental and is subject to change.`
comment to `archiveChat` (the `watchChatDesktop` handler already had it,
just needed the swagger block removed).
Adds `docs/ai-coder/agents/chat-api.md` — a concise guide for the
experimental `/api/experimental/chats` endpoints.
**What's included:**
- Authentication
- Quick start curl example
- Core workflow (create → stream → follow-up)
- All major endpoints: create, messages, stream, list, get, archive,
interrupt
- File uploads
- Chat status reference
Also marks all Coder Agents child pages as `early access` in
`docs/manifest.json`.
The `/chats/{chat}/diff-status` endpoint was redundant because:
- The `Chat` type already has a `DiffStatus` field
- Listing chats already resolves and returns `diff_status`
- The `getChat` endpoint was the only one not resolving it (passing
`nil`)
## Changes
**Backend:**
- `getChat` now calls `resolveChatDiffStatus` and includes the result in
the response
- Removed `getChatDiffStatus` handler, route (`GET /diff-status`), and
SDK method
- Tests updated to use `GetChat` instead of `GetChatDiffStatus`
**Frontend:**
- `AgentDetail.tsx`: uses `chatQuery.data?.diff_status` instead of
separate query
- `RemoteDiffPanel.tsx`: accepts `diffStatus` as a prop instead of
fetching internally
- `AgentsPage.tsx`: `diff_status_change` events now invalidate the chat
query
- Removed `chatDiffStatus` query, `chatDiffStatusKey`, and
`getChatDiffStatus` API method
The release calendar was outdated — it still showed v2.30 as Mainline
and v2.31 as Not Released.
This runs the `scripts/update-release-calendar.sh` script and manually
re-adds the ESR rows that the script doesn't handle:
**Changes:**
- v2.28: Security Support → Not Supported
- v2.29: Stable + ESR → Security Support + ESR (v2.29.8)
- v2.30: Mainline → Stable (v2.30.3)
- v2.31: Not Released → Mainline (v2.31.5)
- Added 2.32 as Not Released
- Kept 2.24 as Extended Support Release
- Updated latest patch versions for all releases
- Removed 2.25 (no longer in the rolling window)
Created on behalf of @matifali
Co-authored-by: blink-so[bot] <211532188+blink-so[bot]@users.noreply.github.com>
Replace the ~370-line bash develop.sh with a Go program using
serpent for CLI flags, errgroup for process lifecycle, and
codersdk for setup. develop.sh becomes a thin make + exec wrapper.
- Process groups for clean shutdown of child trees
- Docker template auto-creation via SDK ExampleID
- Idempotent setup (users, orgs, templates)
- Configurable --port, --web-port, --proxy-port
- Preflight runs lib.sh dependency checks
- TCP dial for port-busy checks
- Make target (build/.bin/develop) for build caching
## Summary
- add an `IS DISTINCT FROM` guard to `InsertChatMessage`'s
`updated_chat` CTE so `chats.last_model_config_id` is only rewritten
when the incoming `model_config_id` actually changes
- regenerate the query layer
- add focused regression coverage for the two meaningful behaviors:
same-model inserts and real model switches
- trim redundant message-field assertions so the new test stays focused
on the guard behavior
## Proof this is an improvement
This PR reduces work in the hottest chat write query without changing
the insert behavior.
### Why the old query did unnecessary work
Before this change, `InsertChatMessage` always ran this update whenever
`model_config_id` was non-null:
```sql
UPDATE chats
SET last_model_config_id = sqlc.narg('model_config_id')::uuid
WHERE id = @chat_id::uuid
AND sqlc.narg('model_config_id')::uuid IS NOT NULL
```
That means the query rewrote the `chats` row even when
`chats.last_model_config_id` was already equal to the incoming value.
### What changes in this PR
This PR adds:
```sql
AND chats.last_model_config_id IS DISTINCT FROM sqlc.narg('model_config_id')::uuid
```
So same-model inserts still insert the message, but they no longer
perform a redundant `UPDATE chats`.
### Why this matters on the hot path
From the chat scaletest investigation that motivated this change:
- `InsertChatMessage` (+ `updated_chat` CTE) was the hottest write query
- about **104k calls**
- about **0.69 ms average latency**
- about **71.8 s total DB execution time**
We also verified common callsites where the update is provably
redundant:
- `CreateChat` inserts the chat with `LastModelConfigID =
opts.ModelConfigID`, then immediately inserts initial system/user
messages with that same model config
- follow-up user messages commonly pass `lockedChat.LastModelConfigID`
straight into `InsertChatMessage`
- assistant/tool/summary persistence keeps the current model in the
common case; only real switches or fallback cases need the chat row
update
That means a meaningful fraction of executions of the hottest DB write
query move from:
- **before:** insert message **+** rewrite chat row
- **after:** insert message only
This should reduce row churn and write contention on `chats`, especially
against other chat-row writers like `UpdateChatStatus` and
`GetChatByIDForUpdate`.
Fixes flaky `TestChatCostSummary_UnpricedMessages` (and siblings) by
replacing implicit handler-default date windows with explicit time
windows derived from database-assigned message timestamps.
**Root cause:** Tests called `GetChatCostSummary` with empty options,
triggering the handler to use `[time.Now()-30d, time.Now())` as the
query window. The SQL filter's exclusive upper bound (`created_at <
@end_date`) can exclude freshly-inserted messages when the handler's
clock drifts even slightly past the message's `created_at`.
**Fix (test-only, `coderd/chats_test.go`):**
- `seedChatCostFixture` now captures `InsertChatMessage` return values
and exposes `EarliestCreatedAt`/`LatestCreatedAt`.
- Added `safeOptions()` helper that builds a padded ±1 min window around
DB timestamps.
- Updated 4 tests to use explicit date windows;
`TestChatCostSummary_DateRange` unchanged.
Validated with `go test -count=20` (100/100 passes).
Adds a new **Enable Coder Agents** section to the Early Access doc
explaining how to activate the `agents` experiment flag via
`CODER_EXPERIMENTS` or `--experiments`.
## Changes
### `docs/ai-coder/agents/early-access.md`
- New **Enable Coder Agents** section with env var and CLI flag
examples.
- Note that the `agents` flag is excluded from wildcard (`*`) opt-in.
- Quick-start checklist: dashboard → Admin → configure provider/model →
start chatting.
- Link to GitHub issues for feedback.
### `docs/ai-coder/agents/index.md`
- Updated **Product status** from "internal preview" to "Early Access"
with a link to the early-access page for enablement instructions.
Instead of showing raw base64 JSON for Anthropic's computer use tool,
render the screenshot as an inline image. The image is clickable to open
at full resolution in a new tab.
## Changes
- **ComputerTool.tsx** — New component that renders base64 image data as
an `<img>` tag
- **Tool.tsx** — Added `ComputerRenderer` handling both single-object
and array-of-blocks result shapes
- **ToolIcon.tsx** — Added `MonitorIcon` for the `computer` tool
- **ToolLabel.tsx** — Added \Screenshot\ label for the `computer` tool
Bumps rust from `d6782f2` to `7d37016`.
[](https://docs.github.com/en/github/managing-security-vulnerabilities/about-dependabot-security-updates#about-compatibility-scores)
Dependabot will resolve any conflicts with this PR as long as you don't
alter it yourself. You can also trigger a rebase manually by commenting
`@dependabot rebase`.
[//]: # (dependabot-automerge-start)
[//]: # (dependabot-automerge-end)
---
<details>
<summary>Dependabot commands and options</summary>
<br />
You can trigger Dependabot actions by commenting on this PR:
- `@dependabot rebase` will rebase this PR
- `@dependabot recreate` will recreate this PR, overwriting any edits
that have been made to it
- `@dependabot show <dependency name> ignore conditions` will show all
of the ignore conditions of the specified dependency
- `@dependabot ignore this major version` will close this PR and stop
Dependabot creating any more for this major version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this minor version` will close this PR and stop
Dependabot creating any more for this minor version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this dependency` will close this PR and stop
Dependabot creating any more for this dependency (unless you reopen the
PR or upgrade to it yourself)
</details>
Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
### Motivation
- The chat creation flow associated a workspace agent for a chat if the requester could read the workspace, enabling privilege escalation where users without SSH/app-connect permissions could cause the daemon to open privileged agent connections and execute commands.
- The intent is to ensure that attaching a workspace agent to a chat only happens when the requester has the workspace SSH permission so the chat daemon cannot be abused to bypass RBAC.
### Description
- Require request-scoped authorization for workspace agent usage by changing `validateCreateChatWorkspaceSelection` to accept the `*http.Request` and calling `api.Authorize(r, policy.ActionSSH, workspace)` before selecting the workspace for a chat.
- Pass the HTTP request into the validator from `postChats` so authorization is evaluated in the request context (`postChats` now calls `validateCreateChatWorkspaceSelection(ctx, r, req)`).
- Add a regression test `WorkspaceAccessibleButNoSSH` in `coderd/chats_test.go` which creates an org-admin-scoped user (read access but no `ActionSSH`) and asserts that creating a chat with `WorkspaceID` is denied.
### Testing
- Ran `gofmt -w coderd/chats.go coderd/chats_test.go` which succeeded.
- Attempted to run repository pre-commit checks (`make pre-commit`) and targeted `go test` invocations; these checks could not be completed in this environment due to missing local tooling and environment constraints (protobuf include resolution, containerized DB access via Docker socket, and long-running golden generation tasks), so full CI/pre-commit verification and end-to-end test runs did not complete here.
- Added a focused regression unit test (`WorkspaceAccessibleButNoSSH`) to prevent reintroduction of the authorization bypass; this test is included in the change and should be executed in CI where the full toolchain and test environment are available.
------
[Codex Task](https://chatgpt.com/codex/tasks/task_b_69b432502670832e91d14e937745de46)
Restore PR title validation that was removed in 828f33a when
cdr-bot was expected to handle it. That bot has since been disabled.
The new title job in contrib.yaml validates:
- Conventional commit format (type(scope): description)
- Type from the same set used by release notes generation
- Scope validity derived from the changed files in the PR diff
- All changed files fall under the declared scope
Uses actions/github-script (no third-party marketplace actions).
Also fixes feat(api) examples across docs (no api folder exists)
and consolidates commit rules into CONTRIBUTING.md as the single
source of truth.
## Why
The DERP health page displayed raw field names like
`MappingVariesByDestIP`, `PMP`, `PCP`, `HairPinning` with no context.
Users without deep networking knowledge had no way to understand what
these flags meant or why they mattered. This change makes the page
self-documenting.
## What
- DERPPage (`/health/derp`)
- Replace flat pill row with four logically grouped tables:
**Connectivity**, **IPv6 Support**, **NAT Traversal**, **Port Mapping**.
- Rename section from "Flags" to "Network Checks".
- Surface `CaptivePortal` flag (previously missing from the UI
entirely).
- Invert display of `MappingVariesByDestIP` and `CaptivePortal` so green
always means good.
- Handle `null` boolean fields (e.g. UPnP, PMP, PCP) with a distinct
"not checked" neutral icon.
- DERPRegionPage (`/health/derp/regions/:regionId`)
- Replace per-node `BooleanPill` row with a table showing **Exchange
Messages**, **Direct HTTP Upgrade**, **STUN Enabled**, and **STUN
Reachable** per node.
- Invert `uses_websocket` display as "Direct HTTP Upgrade" (green when
websocket is not needed).
- Surface **STUN Enabled** and **STUN Reachable** per node (data was
returned by the API but never rendered).
- Add null guards for `region` and `node` (remove `!` non-null
assertions).
- Convert all emotion/MUI styles to Tailwind classes; remove
`reportStyles` object and `useTheme` import.
- Content.tsx (shared)
- Adds an exported `StatusIcon` component with three states: `true`
(green check), `false` (red minus), `null` (neutral help icon).
The script source claimed Dev Containers are early access and told
users to set CODER_AGENT_DEVCONTAINERS_ENABLE=true, which already
defaults to true. Clear the script source and set RunOnStart to
false since there is nothing to run.
Dependabot will resolve any conflicts with this PR as long as you don't
alter it yourself. You can also trigger a rebase manually by commenting
`@dependabot rebase`.
[//]: # (dependabot-automerge-start)
[//]: # (dependabot-automerge-end)
---
<details>
<summary>Dependabot commands and options</summary>
<br />
You can trigger Dependabot actions by commenting on this PR:
- `@dependabot rebase` will rebase this PR
- `@dependabot recreate` will recreate this PR, overwriting any edits
that have been made to it
- `@dependabot show <dependency name> ignore conditions` will show all
of the ignore conditions of the specified dependency
- `@dependabot ignore <dependency name> major version` will close this
group update PR and stop Dependabot creating any more for the specific
dependency's major version (unless you unignore this specific
dependency's major version or upgrade to it yourself)
- `@dependabot ignore <dependency name> minor version` will close this
group update PR and stop Dependabot creating any more for the specific
dependency's minor version (unless you unignore this specific
dependency's minor version or upgrade to it yourself)
- `@dependabot ignore <dependency name>` will close this group update PR
and stop Dependabot creating any more for the specific dependency
(unless you unignore this specific dependency or upgrade to it yourself)
- `@dependabot unignore <dependency name>` will remove all of the ignore
conditions of the specified dependency
- `@dependabot unignore <dependency name> <ignore condition>` will
remove the ignore condition of the specified dependency and ignore
conditions
</details>
Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
Splits the single `coder` artifact (containing all platforms in a 1.3GB
zip) into individual artifacts per OS/arch/format.
## Problem
All CI build artifacts are uploaded as a single artifact named `coder`,
producing a 1.3GB zip containing every platform's binary. This makes it
impossible to download a single platform's binary without pulling the
entire bundle.
## Solution
Upload each platform/format combination as a separate artifact:
| Artifact Name | Contents |
|---|---|
| `coder-linux-amd64.tar.gz` | Linux amd64 tarball |
| `coder-linux-amd64.deb` | Linux amd64 deb package |
| `coder-linux-arm64.tar.gz` | Linux arm64 tarball |
| `coder-linux-arm64.deb` | Linux arm64 deb package |
| `coder-linux-armv7.tar.gz` | Linux armv7 tarball |
| `coder-linux-armv7.deb` | Linux armv7 deb package |
| `coder-windows-amd64.zip` | Windows amd64 zip |
## Plan
This is the first step toward letting customers install directly from
`main` via:
```bash
curl -L https://coder.com/install.sh | sh -s -- --unsafe-unstable
```
GitHub Actions artifact downloads require authentication even for public
repos, so the next steps are to add a small Cloudflare Worker (similar
to the one we already have for `install.sh`) that:
1. Lists artifacts via the GitHub API (unauthenticated) to find the
latest artifact ID for the requested platform
2. Calls the download endpoint with a GitHub token (CF Worker secret) to
get a 302 redirect to a time-limited Azure Blob URL
3. Redirects the caller to that URL (which requires no auth)
This gives us publicly accessible per-platform URLs that the
`--unsafe-unstable` flag would point at. The worker doesn't proxy the
binary itself — it only proxies the metadata API call (~1KB) and
redirects for the actual download.
This PR splits the artifacts so the worker can serve individual platform
downloads (~200MB each) instead of forcing a 1.3GB bundle.
ForkReap's syscall.ForkExec and process-directed signals remain
flaky in CI despite the subprocess isolation added in #22894.
Restore the testutil.InCI() skip guard that was removed in that
change.
Fixescoder/internal#1402
## Summary
Refactors the Git panel in the Agents page to consolidate duplicated
diff viewer code and significantly improve the UI.
### Deduplication
- **RemoteDiffPanel** now uses the shared `DiffViewer` component instead
of duplicating file tree, lazy loading, scroll tracking, and layout
(~500 lines removed).
- Renamed `RepoChangesPanel` → `LocalDiffPanel`, `FilesChangedPanel` →
`RemoteDiffPanel` to reflect actual scope.
- Removed `headerLeft`/`headerRight` abstraction from `DiffViewer` —
each consumer owns its own header.
- Replaced hand-rolled `ChatDiffStatusResponse` with auto-generated
`ChatDiffStatus` from `typesGenerated.ts`.
### Tab Redesign
- Per-repo tabs: each local repo gets its own tab (`Working <repo>`)
instead of a single stacked view.
- PR tab shows state icon + PR title; branch-only tab shows branch icon.
- Tabs use `Button variant="outline"` matching the Git/Desktop tab
style.
- Radix `ScrollArea` with thin horizontal scrollbar for tab overflow.
- Diff style toggle and refresh button lifted to shared toolbar, always
visible.
### PR Header
- Compact sub-header: `base_branch ←`, state badge
(`Open`/`Draft`/`Merged`/`Closed`), diff stats, and `View PR` button.
- GitHub-style state-aware icons (green open, gray draft, purple merged,
red closed).
- New API fields synced: `base_branch`, `author_login`, `pr_number`,
`commits`, `approved`, `reviewer_count`.
### Local Changes Header
- Compact sub-header: branch name, repo root path, diff stats, and
`Commit` button (styled to match `View PR`).
- `CircleDotIcon` (amber) for working changes tabs — universal
"modified" indicator.
### Visual Polish
- All text in sub-headers and buttons at 13px matching chat font size.
- All badges (`DiffStatBadge`, PR state, `View PR`, `Commit`) use
consistent `border-border-default`, `rounded-sm`, `leading-5`.
- No background color on diff viewer header bars.
- Tabs hidden when their view has no content; auto-switch when active
tab disappears.
### Stories
- New `GitPanel.stories.tsx` covering: open PR + working changes, draft
PR, merged PR, closed PR, branch only, working changes only, multiple
repos, empty state.
- Removed old `LocalDiffPanel.stories.tsx` and
`RemoteDiffPanel.stories.tsx`.
## Problem
Sending a message on the `/agents` page triggers a burst of redundant
HTTP requests. The root cause is that chat mutations call
`invalidateQueries({ queryKey: ["chats"] })` which, due to React Query's
default **prefix matching**, cascades to every query whose key starts
with `["chats"]`:
- `["chats", {archived: false}]` — infinite sidebar list
- `["chats", chatId]` — individual chat detail
- `["chats", chatId, "messages"]` — all messages
- `["chats", chatId, "diff-status"]` — diff status
- `["chats", chatId, "diff-contents"]` — diff contents
- `["chats", "costSummary", ...]` — cost summaries
All of these have active subscribers on the page, so each one fires a
network request. The WebSocket stream already delivers these updates in
real-time, making the HTTP refetches completely redundant.
## Fix
| Mutation | Before | After |
|---|---|---|
| `createChatMessage` | `invalidateQueries({ queryKey: chatsKey })` —
prefix cascade | **Removed** — WebSocket delivers messages + sidebar
updates |
| `interruptChat` | `invalidateQueries({ queryKey: chatsKey })` — prefix
cascade | **Removed** — WebSocket delivers status changes |
| `editChatMessage` | 3 broad invalidations including `chatsKey` prefix
| 2 targeted with `exact: true`: `chatKey(id)` + `chatMessagesKey(id)` |
| `promoteChatQueuedMessage` | 3 broad invalidations including
`chatsKey` prefix | 2 targeted with `exact: true`: `chatKey(id)` +
`chatMessagesKey(id)` |
`editChatMessage` keeps `chatMessagesKey` invalidation because editing
truncates messages server-side and the WebSocket can only insert/update,
never remove stale entries.
## Net effect
Sending a message previously triggered **5–7 HTTP requests**. Now it
triggers **zero** — the WebSocket handles everything.
## Tests
Added `describe("mutation invalidation scope")` with 8 test cases
asserting that each mutation only invalidates the queries it genuinely
needs.
## Problem
When creating a new chat in the agents page (`/agents`), the chat could
appear multiple times in the sidebar. This was a race condition
triggered by the WebSocket `created` event handler.
## Root Cause
`updateInfiniteChatsCache` applies its updater function **independently
on each page** of the infinite query:
```ts
const nextPages = prev.pages.map((page) => updater(page));
```
When the `watchChats` WebSocket received a `"created"` event, the
handler checked `exists` only within the *current page*, then prepended
the new chat if not found:
```ts
updateInfiniteChatsCache(queryClient, (chats) => {
const exists = chats.some((c) => c.id === updatedChat.id);
// ...
if (chatEvent.kind === "created") {
return [updatedChat, ...chats]; // runs per page!
}
});
```
Since a brand-new chat doesn't exist in any page, **every loaded page**
prepends it. After `pages.flat()`, the chat appears once per loaded page
in the sidebar.
## Fix
- Added `prependToInfiniteChatsCache` in `chats.ts` that checks across
**all pages** before prepending, and only adds to page 0.
- Split the WebSocket handler so `"created"` events use the new safe
prepend, while update events (`title_change`, `status_change`) continue
using `updateInfiniteChatsCache` (which is safe for `.map()` operations
that don't add entries).
Adds the `head_branch` field (the source/feature branch name of a PR) to
the diff status pipeline. Previously only `base_branch` (target branch)
and the head commit SHA were captured from the GitHub API, but not the
head branch name itself.
## Changes
- **Migration 438**: Add `head_branch` nullable TEXT column to
`chat_diff_statuses`
- **gitprovider**: Parse `head.ref` from the GitHub API response
(alongside `head.sha`) and add `HeadBranch` to `PRStatus`
- **gitsync**: Wire `HeadBranch` through `refreshOne()` into the DB
upsert params
- **worker**: Map `HeadBranch` in `chatDiffStatusFromRow()`
- **coderd**: Convert `HeadBranch` in `convertChatDiffStatus()`
- **codersdk**: Expose as `head_branch` (`*string`, omitempty) in
`ChatDiffStatus` API response
- **Tests**: Updated `github_test.go` pull JSON fixtures and assertions
### Motivation
- The desktop watch handler opened a VNC stream using the chat's
workspace ID while only relying on workspace read permissions, allowing
read-only users to escalate to interactive desktop access.
- Enforce connect-level authorization so only actors with
`ActionApplicationConnect` or `ActionSSH` can open the desktop stream.
### Description
- Added an explicit workspace lookup in `watchChatDesktop` using
`GetWorkspaceByID` to obtain a workspace object for authorization.
- Require the requester to be authorized for either
`policy.ActionApplicationConnect` or `policy.ActionSSH` on the workspace
before proceeding to locate agents or connect to the VNC stream, and
return `403 Forbidden` when neither permission is present.
- The change is minimal and localized to `coderd/chats.go` and does not
alter other code paths or behavior when the requester has the necessary
connect permissions.
### Testing
- Ran `gofmt -w coderd/chats.go` to format the modified file, which
succeeded.
- Attempted to run the unit test `TestWatchChatDesktop/NoWorkspace` via
`go test` in this environment but the test run did not complete within
the environment constraints and did not produce a full pass result.
- Attempted to run the repository pre-commit/gen steps but they could
not complete due to missing developer tooling and services in this
environment (e.g. `sqlc`, `mockgen`, `protoc` plugins and test services
like Docker/Postgres), so full pre-commit validation did not finish
here.
- Code review and static validation confirm the added authorization
check properly prevents read-only access from opening the desktop VNC
stream.
------
[Codex
Task](https://chatgpt.com/codex/tasks/task_b_69b46a4ac5c4832ea9d330aeba43c32d)
## Problem
The agents sidebar infinite scroll was spamming the `/api/v2/chats`
endpoint with duplicate requests at the same offset, caused by the
`LoadMoreSentinel` component.
### Root cause
`onLoadMore` is an inline arrow function (`() => void
chatsQuery.fetchNextPage()`), creating a **new function reference on
every render**. The `useEffect` in `LoadMoreSentinel` depended on
`[onLoadMore]`, so it tore down and re-created the
`IntersectionObserver` on every render. Each new observer immediately
fired its callback when the sentinel was already visible, triggering
duplicate fetches.
## Fix
- Store `onLoadMore` and `isFetchingNextPage` in **refs** so the
observer callback always reads the latest values without needing to tear
down/re-create.
- Create the `IntersectionObserver` **once on mount** (empty deps
array).
- **Guard** against calling `onLoadMore` while `isFetchingNextPage` is
true.
## Tests
- **LoadMoreSentinel behavior tests** (6 tests): verifies no duplicate
calls across re-renders, proper `isFetchingNextPage` gating, ref-based
observer stability, and correct resume after fetch completes.
- **`infiniteChats` query factory tests** (6 tests): covers
`getNextPageParam` and `queryFn` offset computation to prevent
pagination regressions.
Surfaces cache token data in the analytics views and fixes table
spacing.
### Changes
- **Cache token columns**: Added cache read and cache write token counts
to all analytics views (user and admin), from SQL queries through Go SDK
types to the frontend tables and summary cards.
- **Table spacing fix**: Replaced the bare React fragment in
`ChatCostSummaryView` with a `space-y-6` container so the model and chat
breakdown tables no longer overlap.
### Data flow
`chat_messages` table already stores `cache_read_tokens` and
`cache_creation_tokens` (and uses them for cost calculation). This PR
aggregates and displays them alongside input/output tokens in:
- Summary cards (6 cards: Total Cost, Input, Output, Cache Read, Cache
Write, Messages)
- Per-model breakdown table
- Per-chat breakdown table
- Admin per-user table
This PR adds a `WatchAllWorkspaces` function with `watch-all-workspaces`
endpoint, which can be used to listen on a single global pubsub channel
for _all_ workspace build updates, and makes use of it in the autostart
scaletest.
This negates the need to use a workspace watch pubsub channel _per_
workspace, which has auth overhead associated with each call. This is
especially relevant in situations such as the autostart scaletest, where
we need to start/stop a set of workspaces before we can configure their
autostart config. The overhead associated with all the watch requests
skews the scaletest results and makes it harder to reason about the
performance of the autostart feature itself.
The autostart scaletest also no longer generates its own metrics nor
does it wait for all the workspaces to actually start via autostart. We
should update the scaletest dashboard after both PRs are merged to
measure autostart performance via the new metrics.
The new function/endpoint and its usage in the autostart scaletest are
gated behind an experiment feature flag, this is something we should
discuss whether we want to enable the endpoint in prod by default or
not. If so, we can remove the experiment.
---------
Signed-off-by: Callum Styan <callumstyan@gmail.com>
Co-authored-by: Claude Opus 4.5 <noreply@anthropic.com>
Co-authored-by: Callum Styan <callum@coder.com>
Install Google Chrome stable directly from `dl.google.com`. Ubuntu 22.04
ships `chromium-browser` as a snap-only package, which does not work in
Docker containers.
```dockerfile
RUN wget -q https://dl.google.com/linux/direct/google-chrome-stable_current_amd64.deb && \\
apt-get install --yes ./google-chrome-stable_current_amd64.deb && \\
rm google-chrome-stable_current_amd64.deb
```
Verified in a running dogfood workspace:
```
$ google-chrome --version
Google Chrome 146.0.7680.75
```
- Use t.Errorf in chattest non-streaming helpers so encoding
failures fail the test
- Thread testing.TB into writeResponsesAPIStreaming and log
SSE write errors instead of silently dropping them
- Bump createworkspace DB error log from Warn to Error
- Use errors.Join for timeout + output error in execute.go
Add UI components for viewing and managing LLM chat cost analytics.
## Changes
- `UserAnalyticsDialog`: personal cost summary with 30-day date range
- `ChatCostSummaryView`: shared component for cost breakdowns by model
and chat
- `ConfigureAgentsDialog`: admin Usage tab with deployment-wide cost
rollup
- Storybook stories for all new and existing components
- Replace `ModelsSection.test.tsx`, `DashboardLayout.test.tsx`,
`AuditPage.test.tsx` with Storybook stories
- Cost-related API client methods and React Query hooks
- Analytics utilities for formatting microdollar values
Backend: #23036
Implement the backend for the desktop feature for agents.
- Adds a new `/api/experimental/chats/$id/desktop` endpoint to coderd
which exposes a VNC stream from a
[portabledesktop](https://github.com/coder/portabledesktop) process
running inside the workspace
- Adds a new `spawn_computer_use_agent` tool to chatd, which spawns a
subagent that has access to the `computer` tool which lets it interact
with the `portabledesktop` process running inside the workspace
- Adds the plumbing to make the above possible
There's a follow up frontend PR here:
https://github.com/coder/coder/pull/23006
2026-03-13 19:49:34 +01:00
692 changed files with 75065 additions and 21401 deletions
description: "Guide for creating, updating, and following up on pull requests in the Coder repository. Use when asked to open a PR, update a PR, rewrite a PR description, or follow up on CI/check failures."
---
# Pull Request Skill
## When to Use This Skill
Use this skill when asked to:
- Create a pull request for the current branch.
- Update an existing PR branch or description.
- Rewrite a PR body.
- Follow up on CI or check failures for an existing PR.
## References
Use the canonical docs for shared conventions and validation guidance:
- PR title and description conventions:
`.claude/docs/PR_STYLE_GUIDE.md`
- Local validation commands and git hooks: `AGENTS.md` (Essential Commands and
Git Hooks sections)
## Lifecycle Rules
1.**Check for an existing PR** before creating a new one:
```bash
gh pr list --head "$(git branch --show-current)" --author @me --json number --jq '.[0].number // empty'
```
If that returns a number, update that PR. If it returns empty output,
create a new one.
2. **Check you are not on main.** If the current branch is `main` or `master`,
create a feature branch before doing PR work.
3. **Default to draft.** Use `gh pr create --draft` unless the user explicitly
asks for ready-for-review.
4. **Keep description aligned with the full diff.** Re-read the diff against
the base branch before writing or updating the title and body. Describe the
entire PR diff, not just the last commit.
5. **Never auto-merge.** Do not merge or mark ready for review unless the user
explicitly asks.
6. **Never push to main or master.**
## CI / Checks Follow-up
**Always watch CI checks after pushing.** Do not push and walk away.
After pushing:
- Monitor CI with `gh pr checks <PR_NUMBER> --watch`.
- Use `gh pr view <PR_NUMBER> --json statusCheckRollup` for programmatic check
status.
If checks fail:
1. Find the failed run ID from the `gh pr checks` output.
2. Read the logs with `gh run view <run-id> --log-failed`.
3. Fix the problem locally.
4. Run `make pre-commit`.
5. Push the fix.
## What Not to Do
- Do not reference or call helper scripts that do not exist in this
repository.
- Do not auto-merge or mark ready for review without explicit user request.
- Do not push to `origin/main` or `origin/master`.
@@ -113,7 +113,7 @@ Coder emphasizes clear error handling, with specific patterns required:
All tests should run in parallel using `t.Parallel()` to ensure efficient testing and expose potential race conditions. The codebase is rigorously linted with golangci-lint to maintain consistent code quality.
Git contributions follow a standard format with commitmessages structured as `type: <message>`, where type is one of `feat`, `fix`, or `chore`.
Git contributions follow [Conventional Commits](https://www.conventionalcommits.org/en/v1.0.0/). See [CONTRIBUTING.md](docs/about/contributing/CONTRIBUTING.md#commit-messages) for full rules. PR titles are linted in CI.
Format: `type(scope): description`. See [CONTRIBUTING.md](docs/about/contributing/CONTRIBUTING.md#commit-messages) for full rules. PR titles are linted in CI.
Format: `type(scope): message`. See [CONTRIBUTING.md](docs/about/contributing/CONTRIBUTING.md#commit-messages) for full rules. PR titles are linted in CI.
# Build the same triage prompt used by the Tasks API workflow.
TASK_PROMPT=$(cat <<'EOF'
Fix ${ISSUE_URL}
1. Use the gh CLI to read the issue description and comments.
2. Think carefully and try to understand the root cause. If the issue is unclear or not well defined, ask me to clarify and provide more information.
3. Write a proposed implementation plan to PLAN.md for me to review before starting implementation. Your plan should use TDD and only make the minimal changes necessary to fix the root cause.
4. When I approve your plan, start working on it. If you encounter issues with the plan, ask me for clarification and update the plan as required.
5. When you have finished implementation according to the plan, commit and push your changes, and create a PR using the gh CLI for me to review.
EOF
)
# Perform variable substitution on the prompt — scoped to $ISSUE_URL only.
# Using envsubst without arguments would expand every env var in scope
# (including CODER_SESSION_TOKEN), so we name the variable explicitly.
Description:"Timeout for workspace jobs (e.g. build, start).",
Value:serpent.DurationOf(&workspaceJobTimeout),
},
{
Flag:"autostart-build-timeout",
Env:"CODER_SCALETEST_AUTOSTART_BUILD_TIMEOUT",
Default:"15m",
Description:"Timeout for the autostart build to complete. Must be longer than workspace-job-timeout to account for queueing time in high-load scenarios.",
Some files were not shown because too many files have changed in this diff
Show More
Reference in New Issue
Block a user
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
Kayla はな