## Problem
Subagent chats were receiving git context (branch, remote origin, PR
status) from their parent or sibling chats' git operations. When a git
operation triggers external auth, the workspace agent sends `chat_id`
identifying which chat initiated it — but this was broken at two levels:
1. **Agent side:** `CODER_CHAT_ID` was never injected into process
environments. `chatd` sets `Coder-Chat-Id` HTTP headers and the
agent extracts them for process isolation, but never propagated
`CODER_CHAT_ID` to `cmd.Env`. So `gitaskpass` always sent an empty
`chat_id`.
2. **Server side:** `workspaceAgentsExternalAuth` ignored the `chat_id`
query param. `MarkStale` broadcast git context to **all** chats on
the workspace via `filterChatsByWorkspaceID`.
## Fix
- Inject `CODER_CHAT_ID` into `cmd.Env` in `agentproc` when the chat
ID is known, so `gitaskpass` can read and forward it.
- Read `chat_id` from query params in `workspaceAgentsExternalAuth`
and thread it through `chatGitRef`.
- Refactor `MarkStale` to accept a `MarkStaleParams` struct. When
`ChatID` is provided, target only that specific chat. When empty
(legacy agents, non-chat git operations), fall back to the existing
workspace-wide broadcast.
- Extract `markStaleSingle` helper to deduplicate the upsert+publish
logic.
<details><summary>Investigation notes</summary>
### Data flow before fix
```
chatd → sets Coder-Chat-Id header on agent conn
agent → extracts chatID, stores on process struct
agent → does NOT set CODER_CHAT_ID in cmd.Env ← gap 1
gitaskpass → reads CODER_CHAT_ID (always empty), sends chat_id=""
server handler → ignores chat_id query param ← gap 2
MarkStale → broadcasts to ALL workspace chats
```
### Data flow after fix
```
chatd → sets Coder-Chat-Id header on agent conn
agent → extracts chatID, stores on process struct
agent → sets CODER_CHAT_ID in cmd.Env
gitaskpass → reads CODER_CHAT_ID, sends chat_id=<uuid>
server handler → reads chat_id, passes to MarkStale
MarkStale → targets only that specific chat
```
</details>
Unarchiving a root chat now restores descendant chats in the database
and emits lifecycle events for every affected chat so passive sessions
converge without a full refetch.
This keeps archive and unarchive symmetric at both the data and
watch-stream layers by returning the affected chat family from the
database, using those post-update rows for chatd pubsub fanout, and
covering descendant lifecycle delivery with a watch-level regression
test.
Closes#23666
Fixes flaky `TestOpenAIReasoningWithWebSearchRoundTripStoreFalse` and
`TestOpenAIReasoningWithWebSearchRoundTrip`.
## Changes
- Gate the `processChat` control subscriber's cancel callback behind a
`chan struct{}` that is closed after publishing `"running"` status
- Add `TestGatedControlCancel` with 4 subtests exercising the gate logic
<details>
<summary>Root cause analysis</summary>
`SendMessage` publishes a `"pending"` notification on
`chat:stream:<chatID>` via PostgreSQL `NOTIFY`. `processChat` subscribes
to the same channel for control signals. Due to async NOTIFY delivery,
the `"pending"` notification can arrive at the control subscriber
**after** it registers its queue — even though it was published
**before**. `shouldCancelChatFromControlNotification("pending")` returns
`true`, immediately self-interrupting the processor before it does any
work.
The fix gates the cancel callback behind a closed channel. The channel
is closed after `processChat` publishes `"running"` status, so stale
notifications from before initialization are harmlessly ignored.
`close()` provides a happens-before guarantee in the Go memory model.
</details>
> 🤖 Written by a Coder Agent. Reviewed by a human.
Archiving a chat now transitions pending or running chats to waiting
before setting the archived flag. This publishes a status notification
on `ChatStreamNotifyChannel` so `subscribeChatControl` cancels the
active `processChat` context via `ErrInterrupted` — the same codepath
used by the stop button.
The `processChat` cleanup also skips queued-message auto-promotion when
the chat is archived, so archiving behaves like a hard stop rather than
interrupt-and-continue.
Relates to https://github.com/coder/coder/issues/23666
The flaky test assumed the second streamed OpenAI request had already
been captured when the chat status event arrived. In practice, the
capture server can record that second request slightly later, which
intermittently left `streamRequestCount` at `1`.
This change waits for the second captured request before asserting on
the follow-up payload and relaxes the count check to a sanity check. The
test still verifies the `store=false` round-trip behavior without
depending on that timing race.
Fixescoder/internal#1433
## Summary
Fixes three flaky chatd tests that intermittently fail due to timing
races with the background run loop.
Closescoder/internal#1428
## Root Cause
`CreateChat` and `PromoteQueued` call `signalWake()` which writes to
`wakeCh`, triggering `processOnce` immediately. Even though
`newTestServer` sets `PendingChatAcquireInterval: testutil.WaitLong` to
prevent ticker-based polling, the wake channel bypasses this. This
causes `processOnce` to acquire and process the chat concurrently with
the test's manual DB updates and assertions.
### Failing tests
| Test | Failure | Cause |
|------|---------|-------|
| `TestPromoteQueuedAllowsAlreadyQueuedMessageWhenUsageLimitReached` |
`expected: "pending", actual: "running"` | Wake from `CreateChat` races
with manual `UpdateChatStatus`; wake from `PromoteQueued` acquires the
chat before the status assertion |
| `TestSendMessageInterruptBehaviorQueuesAndInterruptsWhenBusy` |
`should have 1 item(s), but has 2` | Wake from `CreateChat` triggers
`processChat` which auto-promotes a queued message, adding an extra row
to `chat_messages` |
| `TestSubscribeNoPubsubNoDuplicateMessageParts` | `Condition satisfied`
(duplicate events) | Pre-existing `WaitGroup.Add/Wait` race in the
`Eventually` + `WaitUntilIdleForTest` pattern |
## Fix
Introduces a `waitForChatProcessed` helper that:
1. Polls until the chat reaches a **terminal state** (not pending AND
not running)
2. Then calls `WaitUntilIdleForTest` to wait for the inflight
`WaitGroup`
Waiting for a terminal state (not just "not pending") avoids a
`sync.WaitGroup` `Add/Wait` race: `AcquireChats` updates the DB status
to `running` **before** `processOnce` calls `inflight.Add(1)`. Checking
only `status != pending` could return while `Add(1)` hasn't happened
yet, causing `Wait()` to return prematurely.
### Per-test changes
- **`TestSendMessageInterruptBehaviorQueuesAndInterruptsWhenBusy`**:
Call `waitForChatProcessed` after `CreateChat` before manually setting
running status
-
**`TestPromoteQueuedAllowsAlreadyQueuedMessageWhenUsageLimitReached`**:
Call `waitForChatProcessed` after `CreateChat`; remove the inherently
racy `status == pending` assertion after `PromoteQueued` (the wake
immediately acquires the chat). Key assertions on promoted message,
queue state, and message count remain.
- **`TestSubscribeNoPubsubNoDuplicateMessageParts`**: Replace inline
`Eventually` with the safer `waitForChatProcessed` helper
## Verification
All three tests pass 150 consecutive executions with `-race -count=10`
across 15 runs (0 failures).
Adds a nullable JSONB column `last_injected_context` to the `chats`
table that stores the most recently persisted injected context parts
(AGENTS.md context-file and skill message parts). The column is updated
only when `persistInstructionFiles()` runs — on first workspace attach
or when the agent changes — so there are no redundant writes on
subsequent turns.
Internal fields (`ContextFileContent`, `ContextFileOS`,
`ContextFileDirectory`, `SkillDir`) are stripped at write time so the
column only holds small metadata. No stripping needed on the read path.
<details>
<summary>Implementation notes</summary>
- New migration `000456` adds nullable `last_injected_context JSONB`
column.
- New SQL query `UpdateChatLastInjectedContext` writes the column
without touching `updated_at`.
- `persistInstructionFiles()` strips internal fields from parts via
`StripInternal()` before persisting.
- Sentinel path (no AGENTS.md) persists skill-only parts when skills
exist.
- `codersdk.Chat` exposes `LastInjectedContext []ChatMessagePart`
(omitempty).
- `db2sdk.Chat()` passes through the already-clean data.
</details>
Adds suffix-based agent selection for chatd. Template authors can direct
chat traffic to a specific root workspace agent by naming it with the
`-coderd-chat` suffix (for example, `coder_agent "dev-coderd-chat"`).
When no suffix match exists, chatd falls back to the first root agent by
`DisplayOrder`, then `Name`. Multiple suffix matches return an error.
The selection logic lives in `coderd/x/chatd/internal/agentselect` and
is shared by chatd core plus the workspace chat tools so all chat entry
points pick the same agent deterministically.
No database migrations, API contract changes, or provider changes. The
experimental sandbox template was split out to #23777.
This fixes a flaky `TestConfigCache_UserPrompt_ExpiredEntryRefetches` by
making the seeded user prompt entry unambiguously expired before the
cache lookup runs.
The test previously inserted a `tlru` entry with a zero TTL, which
depends on `Set` and `Get` landing in different clock ticks. Switching
that seed entry to a negative TTL keeps the bounded `tlru` cache
behavior while removing the same-tick race.
Close https://github.com/coder/internal/issues/1432
## Summary
Skills are now discovered once on the first turn (or when the workspace
agent changes) and persisted as `skill` message parts alongside
`context-file` parts. On subsequent turns, the skill index is
reconstructed from persisted parts instead of re-dialing the workspace
agent.
This makes skills consistent with the AGENTS.md pattern and is
groundwork for a future `/context` endpoint that surfaces loaded
workspace context to the frontend.
## Changes
- Add `skill` `ChatMessagePartType` with `SkillName` and
`SkillDescription` fields
- Extend `persistInstructionFiles` to also discover and persist skills
as parts
- Add `skillsFromParts()` to reconstruct skill index from persisted
parts on subsequent turns
- Update `runChat()` to use `skillsFromParts` instead of re-dialing
workspace for skills
- Frontend: handle new `skill` part type (skip rendering, hide
metadata-only messages)
## Before / After
| | AGENTS.md | Skills |
|---|---|---|
| **Before** | Persist as `context-file` parts, reconstruct from parts |
In-memory `skillsCache` only, re-dial workspace on cache miss |
| **After** | Persist as `context-file` parts, reconstruct from parts |
Persist as `skill` parts, reconstruct from parts |
The in-memory `skillsCache` remains for `read_skill`/`read_skill_file`
tool calls that need full skill bodies on demand.
<details><summary>Design context</summary>
This is the first step toward a unified workspace context
representation. Currently:
- Context files are persisted as message parts (works)
- Skills were only in-memory (inconsistent)
- Workspace MCP servers are cached in-memory (future work)
Persisting skills as parts means a future `/context` endpoint can query
both context files and skills from the same message parts in the DB,
without depending on ephemeral server-side caches.
</details>
Previously, generating a new agent title used a page-global pending
state, so one in-flight regeneration disabled the action for every chat
in the Agents UI.
This change tracks regenerations by chat ID, updates the Agents page
contracts to use `regeneratingTitleChatIds`, and adds sidebar story
coverage that proves only the active chat is disabled.
Previously, when a user sent a message, there was a 0–1000ms (avg
~500ms) polling delay before processing began.
`SendMessage`/`CreateChat`/`EditMessage` set `status='pending'` in the
DB and returned, but nothing woke the processing loop — it was a blind
1-second ticker.
## Changes
**Event-driven acquisition (main change):** Adds a `wakeCh` channel to
the chatd `Server`. `CreateChat`, `SendMessage`, `EditMessage`, and
`PromoteQueued` call `signalWake()` after committing their transactions,
which wakes the run loop to call `processOnce` immediately. The 1-second
ticker remains as a fallback safety net for edge cases (stale recovery,
missed signals).
**Buffer WebSocket write channel:** Changes the
`OneWayWebSocketEventSender` event channel from unbuffered to buffered
(64), decoupling the event producer from WebSocket write speed. The
existing 10s write timeout guards against stuck connections.
<details><summary>Implementation plan & analysis</summary>
The full latency analysis identified these sources of delay in the
streaming pipeline:
1. **Chat acquisition polling** — 0–1000ms (avg 500ms) dead time per
message. Fixed by wake channel.
2. **Unbuffered WebSocket write channel** — each token blocked on the
previous WS write completing. Fixed by buffering.
3. **PersistStep DB transaction per step** — `FOR UPDATE` lock + batch
insert. Not addressed in this PR (medium risk, would overlap DB write
with next provider TTFB).
4. **Multi-hop channel pipeline** — 4 channel hops per token. Not
addressed (medium complexity).
</details>
<details><summary>Test stabilization notes</summary>
`signalWake()` causes the chatd daemon to process chats immediately
after creation/send/edit, which exposed timing assumptions in several
tests that expected chats to remain in `pending` status long enough to
assert on. These tests were updated with `require.Eventually` +
`WaitUntilIdleForTest` patterns to wait for processing to settle before
asserting.
The race detector (`test-go-race-pg`) shows failures in
`TestCreateWorkspaceTool_EndToEnd` and `TestAwaitSubagentCompletion` —
these appear to be pre-existing races in the end-to-end chat flow that
are now exercised more aggressively because processing starts
immediately instead of after a 1s delay. Main branch CI (race detector)
passes without these changes.
</details>
Short prompts were producing title-generation meta responses such as "I
am a title generator" and prompt-echo titles. This rewrites the
automatic and manual title prompts to be shorter, less self-referential,
and more focused on returning only the title text.
The change also removes the broader post-generation guard layer, updates
manual regeneration to send real conversation text instead of a meta
instruction, and keeps regression coverage focused on the slimmer prompt
contract.
Adds skill discovery and tools to chatd so the agent can discover and
load `.agents/skills/` from workspaces, following the same pattern as
AGENTS.md instruction loading and MCP tool discovery.
## What changed
### `chattool/skill.go` — discovery, loading, and tools
- **DiscoverSkills** — walks `.agents/skills/` via `conn.LS()` +
`conn.ReadFile()`, parses SKILL.md frontmatter (name + description),
validates kebab-case names match directory names, silently skips
broken/missing entries.
- **FormatSkillIndex** — renders a compact `<available-skills>` XML
block for system prompt injection (~60 tokens for 3 skills). Progressive
disclosure: only names + descriptions in context, full body loaded on
demand.
- **LoadSkillBody** / **LoadSkillFile** — on-demand loading with path
traversal protection and size caps (64KB for SKILL.md, 512KB for
supporting files).
- **read_skill** / **read_skill_file** tools — `fantasy.AgentTool`
implementations following the same pattern as ReadFile and
WorkspaceMCPTool. Receive pre-discovered `[]SkillMeta` via closure to
avoid re-scanning on every call.
### `chatd.go` — integration into runChat
- Skills discovered in the `g2` errgroup parallel with instructions and
MCP tools.
- `skillsCache` (sync.Map) per chat+agent, same invalidation pattern as
MCP tools cache.
- Skill index injected via `InsertSystem` after workspace instructions.
- Re-injected in `ReloadMessages` callback so it survives compaction.
- `read_skill` + `read_skill_file` tools registered when skills are
present (for both root and subagent chats).
- Cache cleaned up in `cleanupStreamIfIdle` alongside MCP tools cache.
## Format compatibility
Uses the same `.agents/skills/<name>/SKILL.md` format as
[coder/mux](https://github.com/coder/mux) and
[openai/codex](https://github.com/openai/codex).
## Summary
Adds read/unread tracking for chats so users can see which agent
conversations have new assistant messages they haven't viewed.
## Backend Changes
- Adds `last_read_message_id` column to the `chats` table (migration
000439).
- Computes `has_unread` as a virtual column in `GetChatsByOwnerID` using
an `EXISTS` subquery checking for assistant messages beyond the read
cursor.
- Exposes `has_unread` on the `codersdk.Chat` struct and auto-generated
TypeScript types.
- Updates `last_read_message_id` on stream connect/disconnect in
`streamChat`, avoiding per-message API calls during active streaming.
- Uses `context.WithoutCancel` for the deferred disconnect write so the
DB update succeeds even after the client disconnects.
## Frontend Changes
- Bold title (`font-semibold`) for unread chats in the sidebar.
- Small blue dot indicator next to the relative timestamp.
- Suppresses unread indicator for the currently active chat via
`isActive` from NavLink.
## Design Decisions
- Only `assistant` messages count as unread — the user's own messages
don't trigger the indicator.
- No foreign key on `last_read_message_id` since messages can be deleted
(via rollback/truncation) and the column is just a high-water mark.
- Zero API calls during streaming: exactly 2 DB writes per stream
session (connect + disconnect).
- Unread state refreshes on chat list load and window focus. The
`watchChats` WebSocket optimistically marks non-active chats as unread
on `status_change` events, but does not carry a server-computed
`has_unread` field. Navigating to a chat optimistically clears its
unread indicator in the cache.
Add a per-MCP-server `model_intent` toggle that wraps tool schemas with
a
`model_intent` field, requiring the LLM to provide a human-readable
description of each tool call's purpose. The intent string is shown as a
status label in the UI instead of opaque tool names, and is
transparently
stripped before the call reaches the remote MCP server.
Built-in tools have rich specialized renderers (terminal blocks, file
diffs,
etc.) and don't need this. MCP tools hit `GenericToolRenderer` which
only
shows raw tool names and JSON — that's where model_intent adds value.
The model learns what to provide via the JSON Schema `description` on
the
`model_intent` property itself — no system prompt changes needed.
<details>
<summary>Implementation details</summary>
### Architecture
Inspired by the `withModelIntent()` pattern from `coder/blink`, adapted
for
Go + React. The wrapping is entirely in the `mcpclient` layer — tool
implementations never see `model_intent`.
**Schema wrapping** (`mcpToolWrapper.Info()`): When enabled, wraps the
original tool parameters under a `properties` key and adds a
`model_intent`
string field with a rich description that teaches the model inline.
**Input unwrapping** (`mcpToolWrapper.Run()`): Strips `model_intent` and
unwraps `properties` before forwarding to the remote MCP server. Handles
three input shapes models may produce:
1. `{ model_intent, properties: {...} }` — correct format
2. `{ model_intent, key: val, ... }` — flat, no wrapper
3. Malformed — falls through gracefully
**Frontend extraction**: `streamState.ts` extracts `model_intent` from
incrementally parsed streaming JSON. `messageParsing.ts` extracts it
from
persisted tool call args.
**UI rendering**: `GenericToolRenderer` shows the capitalized intent
string
as the primary label when available, falling back to the raw tool name.
### Changes
- Database: `model_intent` boolean column on `mcp_server_configs`
- SDK: `ModelIntent` field on config/create/update types
- API: pass-through in create/update handlers + converter
- mcpclient: schema wrapping in `Info()`, input unwrapping in `Run()`
- Frontend: extraction from streaming + persisted args
- UI: intent label in `GenericToolRenderer`, toggle in admin panel
- Tests: 6 new tests (schema wrapping, unwrapping, passthrough,
fallback)
### Decision log
- **Option lives on MCPServerConfig, not model config**: Built-in tools
already have rich renderers; only MCP tools benefit from model_intent.
- **No system prompt changes**: The JSON Schema `description` on the
`model_intent` property teaches the model inline.
- **Pointer bool on update request**: Follows existing pattern (`*bool`)
so PATCH requests don't reset the value when omitted.
</details>
Fixes expired MCP OAuth2 tokens causing 401 errors and stale
`auth_connected` status in the UI.
When users authenticate MCP servers (e.g. GitHub) via OAuth2, the access
token and refresh token are stored in the database. However, when the
access token expired, nothing refreshed it anywhere:
- **chatd**: sent the expired token as-is, getting a 401 and skipping
the MCP server
- **list/get endpoints**: reported `auth_connected: true` just because a
token record existed, regardless of expiry
## Changes
### Shared utility: `mcpclient.RefreshOAuth2Token`
Pure function that uses `golang.org/x/oauth2` `TokenSource` to check if
a token is expired (or within 10s of expiry) and refresh it. No DB
dependency — callers handle persistence.
### chatd (`coderd/x/chatd/chatd.go`)
Before calling `mcpclient.ConnectAll`, refreshes expired tokens.
Persists new credentials to the database. Falls back to the old token if
refresh fails.
### List/get MCP server endpoints (`coderd/mcp.go`)
Both `listMCPServerConfigs` and `getMCPServerConfig` now attempt refresh
when checking `auth_connected`. If the token is expired:
- **Has refresh token**: attempt refresh, persist result, report
`auth_connected` based on success
- **No refresh token**: report `auth_connected: false` if expired
This means the UI accurately reflects whether the user's token is
actually usable, rather than just whether a record exists.
<details>
<summary>Design notes</summary>
- `RefreshOAuth2Token` lives in `mcpclient` to avoid circular imports
(`coderd` → `chatd` → `mcpclient` is fine; `chatd` → `coderd` would be
circular).
- DB persistence is handled by each caller with their own authz context
(`AsSystemRestricted` in both cases).
- The `buildAuthHeaders` warning in mcpclient about expired tokens is
kept as defense-in-depth logging.
</details>
## Problem
Chats with a persisted `agent_id` binding hang indefinitely when the
workspace is stopped. The stale agent row still exists in the DB, so
`ensureWorkspaceAgent` succeeds, but the dial blocks forever in
`AwaitReachable`. The MCP discovery goroutine used an unbounded context,
so `g2.Wait()` never returned and the LLM never started.
## Fix
Three targeted changes restore the pre-binding behavior where stopped
workspaces degrade gracefully instead of blocking:
1. **`dialWithLazyValidation`**: "no agents in latest build" is now a
terminal fast-fail — the hanging dial is canceled and
`errChatHasNoWorkspaceAgent` returned immediately, instead of falling
through to `waitForOriginalDial`.
2. **Pre-LLM workspace setup**: MCP discovery and instruction
persistence gate on `workspaceAgentIDForConn` before attempting any
dial. MCP discovery is bounded by a 5s timeout and checks the in-memory
tool cache first (using the cheap cached agent from
`ensureWorkspaceAgent`), so the common subsequent-turn path has zero DB
queries.
3. **`persistInstructionFiles`**: tracks whether the workspace
connection succeeded and skips sentinel persistence on failure, so the
next turn retries if the workspace is restarted.
## Scenarios
**Running workspace, subsequent turn (hot path):** MCP cache hit via
in-memory cached agent. Zero DB queries, zero dials. Unchanged from
#23274.
**Stopped workspace, persisted binding (the bug):** MCP cache hit (stale
descriptors, fine — they fail at invocation). Pre-LLM setup completes
instantly. Tool invocation enters `dialWithLazyValidation`, dial fails
or hangs, validation discovers no agents, returns
`errChatHasNoWorkspaceAgent`. Model sees the error and can call
`start_workspace`.
**New chat, running workspace:** `ensureWorkspaceAgent` resolves via
latest-build, persists binding. MCP discovery dials and caches tools.
**New chat, stopped workspace:** `ensureWorkspaceAgent` finds no agents,
returns `errChatHasNoWorkspaceAgent`. Pre-LLM setup skips. LLM starts
with built-in tools only.
**Rebuilt workspace (agent switched):** MCP cache hit with stale agent
(harmless for one turn). Tool invocation dials stale agent, fails fast,
`dialWithLazyValidation` switches to new agent, persists updated
binding.
**Workspace restarted after stop:** No sentinel was persisted during the
stopped turn, so instruction persistence retries. Agent binding switches
to the new agent via `workspaceAgentIDForConn`.
**Transient DB error during validation:** Not
`errChatHasNoWorkspaceAgent`, so `dialWithLazyValidation` falls through
to `waitForOriginalDial` (cannot prove stale). No false positive.
**Tool invocation on stopped workspace:** `getWorkspaceConn` calls
`ensureWorkspaceAgent` (returns stale row), then
`dialWithLazyValidation` validation discovers no agents, returns
`errChatHasNoWorkspaceAgent`, cached state cleared, error returned to
model.
## Problem
The agent chat delayed-startup banner ("Response startup is taking
longer than expected") could appear even though the model was already
streaming.
The root cause is in `Subscribe()`: `message_part` events were delivered
via the fast local in-process stream, while `status` events were
delivered via PostgreSQL pubsub. Both feed into the same `select`
statement, and Go's `select` picks whichever channel is ready first —
there is no ordering guarantee between channels. So a `message_part`
could outrun the `status=running` that logically precedes it.
The frontend saw content arrive while it still thought the chat was
pending, triggering the banner.
## Fix
Also forward `status` events from the local channel, alongside
`message_part`.
Both event types already travel through the same FIFO subscriber
channel: `publishStatus()` is called before the first `message_part`, so
channel ordering guarantees the frontend sees `status=running` before
any content.
Pubsub still delivers a duplicate `status` event later; the frontend
deduplicates it (`setChatStatus` is idempotent — it early-returns when
the status hasn't changed).
## Summary
Adds a "Generate new title" action that lets users manually regenerate a
chat's title using richer conversation context than the automatic
first-message title path.
## Changes
### Backend
- **New endpoint:** `POST
/api/experimental/chats/{chatID}/title/regenerate` returns the updated
Chat with a regenerated title
- **Manual title algorithm:** Extracts useful user/assistant text turns
→ selects first user turn + last 3 turns → builds context with gap
markers → renders prompt with anti-recency guidance → calls lightweight
model → normalizes output
- **Helpers:** `extractManualTitleTurns`,
`selectManualTitleTurnIndexes`, `buildManualTitleContext`,
`renderManualTitlePrompt`, `generateManualTitle` — all private, with the
public `Server.RegenerateChatTitle` method
- **SDK:** `ExperimentalClient.RegenerateChatTitle(ctx, chatID) (Chat,
error)`
- Persists title via existing `UpdateChatByID` and broadcasts
`ChatEventKindTitleChange`
### Frontend
- API client method + React Query mutation with cache invalidation
- "Generate new title" menu item (with wand icon) in both TopBar and
Sidebar dropdown menus
- Loading/disabled state while regeneration is in-flight
- Error toast on failure
- Stories updated for both menus
### Tests
- `quickgen_test.go`: Table-driven tests for all 4 helper functions
(turn extraction, index selection, context building, prompt rendering)
- `exp_chats_test.go`: Handler tests (ChatNotFound,
NotFoundForDifferentUser, NoDaemon)
## Design notes
- The existing auto-title path (`maybeGenerateChatTitle`, `titleInput`)
is completely unchanged
- Manual regeneration uses richer context (first user turn + last 3
turns + gap markers) vs the auto path's single first message
- Endpoint is experimental and marked with `@x-apidocgen {"skip": true}`
Coder's chat (chatd) can now discover and use MCP servers configured in
a workspace's `.mcp.json` file. This brings project-specific tooling
(GitHub, databases, docs servers, etc.) into the chat without any manual
configuration.
## How it works
The workspace agent reads `.mcp.json` from the workspace directory (same
format Claude Code uses), connects to the declared MCP servers —
spawning child processes for stdio servers and connecting over the
network for HTTP/SSE — and caches their tool lists. Two new agent HTTP
endpoints expose this:
- `GET /api/v0/mcp/tools` returns the cached tool list (supports
`?refresh=true`)
- `POST /api/v0/mcp/call-tool` proxies calls to the correct server
On each chat turn, chatd calls `ListMCPTools` through the existing
`AgentConn` tailnet connection, wraps each tool as a
`fantasy.AgentTool`, and adds them to the LLM's tool set alongside
built-in and admin-configured MCP tools. Tool names are prefixed with
the server name (`github__create_issue`) to avoid collisions.
Failed server connections are logged and skipped — they never block the
agent or break the chat. Child stdio processes are terminated on agent
shutdown.
*Problem:* `publishChatPubsubEvent` was constructing a partial
`codersdk.Chat` that omitted `LastModelConfigID` and other fields. Go's
zero-value UUID caused the sidebar to show "Default model" for chats
received via SSE.
*Solution:*
- Extracted `convertChat`/`convertChats` from `exp_chats.go` into
`db2sdk.Chat`/`db2sdk.Chats`, alongside existing `ChatMessage`,
`ChatQueuedMessage`, and `ChatDiffStatus` converters.
`publishChatPubsubEvent` now calls `db2sdk.Chat(chat, nil)` instead of
maintaining its own copy of the conversion logic
- Added backend integration test
`TestWatchChats/CreatedEventIncludesAllChatFields`
- Added frontend regression tests for nil-UUID and valid model config ID
cases
> 🤖 Created by Coder Agents, reviewed by this human.
> **PR Stack**
> 1. **#23351** ← `#23282` *(you are here)*
> 2. #23282 ← `#23275`
> 3. #23275 ← `#23349`
> 4. #23349 ← `main`
---
## Summary
`chatretry.Retry()` used pure exponential backoff (1 s, 2 s, 4 s, …) and
never consulted provider `Retry-After` headers. Fantasy's
`ProviderError` carries `ResponseHeaders` including `Retry-After`, but
`chaterror.Classify()` only parsed error text and silently dropped the
structured transport metadata.
This makes `Retry-After` a first-class signal in the classification →
retry pipeline.
<img width="853" height="346" alt="image"
src="https://github.com/user-attachments/assets/65f012b6-8173-43d2-957e-ab9faddea525"
/>
## Changes
### `coderd/chatd/chaterror/classify.go`
- Added `RetryAfter time.Duration` field to `ClassifiedError` — a
normalized minimum retry delay derived from provider response metadata.
- `Classify()` now calls `extractProviderErrorDetails()` before falling
back to text heuristics. Structured `ProviderError.StatusCode` takes
priority over regex extraction.
- `normalizeClassification()` preserves and clamps `RetryAfter`.
### `coderd/chatd/chaterror/provider_error.go` (new)
Provider-specific extraction, isolated from the text-based
classification logic:
- `extractProviderErrorDetails()` unwraps `*fantasy.ProviderError` from
the error chain via `errors.As`.
- `retryAfterFromHeaders()` parses headers in priority order:
1. `retry-after-ms` (OpenAI-specific, millisecond precision)
2. `retry-after` (standard HTTP — integer seconds or HTTP-date)
- Case-insensitive header key lookup.
### `coderd/chatd/chatretry/chatretry.go`
- `effectiveDelay(attempt, classified)` computes `max(Delay(attempt),
classified.RetryAfter)` — the provider hint acts as a floor without
weakening the local exponential backoff.
- `Retry()` now uses `effectiveDelay` and passes the effective delay to
both `onRetry(...)` and the sleep timer, so downstream payloads, logs,
and the frontend countdown stay aligned automatically.
### Tests
- `classify_test.go`: Structured provider status + `Retry-After`
extraction, `retry-after-ms` priority, HTTP-date parsing, invalid header
fallback, `WithProvider` preservation.
- `chatretry_test.go`: Retry-after-as-floor semantics — longer hint
wins, shorter hint keeps base delay.
## Design notes
- **No SDK/API/frontend changes needed.** `codersdk.ChatStreamRetry`
already carries `DelayMs` and `RetryingAt`, and the frontend already
consumes them. The fix is purely in the server-side delay computation.
- **Existing retryability rules unchanged.** This fixes *when* we sleep,
not *whether* an error is retryable.
- **Provider hint is a floor:** `max(baseDelay, RetryAfter)` ensures we
never retry earlier than the provider asks, and never weaken our own
backoff curve.
## Changes
- **Commit 1**: Remove 17 unnecessary `//nolint` directives:
- `//nolint:varnamelen` — linter not active
- `//nolint:unused` on exported `SlimUnsupported`
- `//nolint:govet` in `coderd/httpmw/csrf` — no longer fires
- `//nolint:revive` on functions refactored since the nolint was added
- `//nolint:paralleltest` citing Go 1.22 loop variable capture
(obsolete)
- Bare `//nolint` narrowed to specific `//nolint:gocritic` with
justification
- **Commit 2**: Fix root causes behind 5 dangerous nolint suppressions:
- Add `MinVersion: tls.VersionTLS12` to TLS client config (removes
`gosec` G402)
- Delete trivial unexported wrappers `apiKey()`/`normalizeProvider()` in
chatprovider (removes `revive` confusing-naming)
- Add doc comments to `StartWithAssert` and `Router` (removes `revive`
exported)
- Rename unused parameters to `_` in integration test helpers
> 🤖 This PR was created using Coder Agents and reviewed by me.
## Summary
Adds a process-wide cache for three hot database queries in `chatd` that
were hitting Postgres on **every chat turn** despite returning
rarely-changing configuration data:
| Query | Before (50k turns) | After | Reduction |
|---|---|---|---|
| `GetEnabledChatProviders` | ~98.6k calls | ~500-1000 | ~99% |
| `GetChatModelConfigByID` | ~49.2k calls | ~500-1000 | ~98% |
| `GetUserChatCustomPrompt` | ~46.7k calls | ~1000-2000 | ~97% |
These were identified via `coder exp scaletest chat` (5000 concurrent
chats × 10 turns) as the dominant source of Postgres load during chat
processing.
## Design
Follows the established **webpush subscription cache pattern**
(`coderd/webpush/webpush.go`):
- `sync.RWMutex` + `tailscale.com/util/singleflight` (generic) +
generation-based stale prevention + TTL
- 10s TTL for provider/model config, 5s TTL for user prompts
- Negative caching for `sql.ErrNoRows` on user prompts (the common case
— most users don't set custom prompts)
- Deep-clones `ChatModelConfig.Options` (`json.RawMessage` = `[]byte`)
on both store and read paths
### Invalidation
Single pubsub channel (`chat:config_change`) with kind discriminator for
cross-replica cache invalidation. Seven publish points in
`coderd/chats.go` cover all admin mutation endpoints
(create/update/delete for providers and model configs, put for user
prompts).
_This PR was generated with mux and was reviewed by a human_
Follow-up to #23282. The retry and terminal error callouts had a few UX
oddities:
- Auto-retrying states reused backend error text that said "Please try
again" even while the UI was already retrying on behalf of the user.
- Terminal error states also said "Please try again" with no action the
user could take.
- `startup_timeout` had no specific title or retry copy — it fell
through to the generic "Retrying request" heading.
- The kind pill showed raw enum values like `startup_timeout` and
`rate_limit`.
- Terminal error metadata showed a "Retryable" / "Not retryable" label
that does not help users.
- A separate "Provider anthropic" metadata row duplicated information
already present in the message body.
- The `usage-limit` error kind used a hyphen while every backend kind
uses underscores.
Changes:
**Backend (`chaterror/message.go`)**
- Split message generation into `terminalMessage()` and
`retryMessage()`, replacing the old `userFacingMessage()`.
- Terminal messages include HTTP status codes and actionable guidance
(e.g. "Check the API key, permissions, and billing settings.").
- Retry messages are clean factual statements without status codes or
remediation, suitable for the retry countdown UI (e.g. "Anthropic is
temporarily overloaded.").
- Removed "Please try again" / "Please try again later" from all paths.
- `StreamRetryPayload` calls `retryMessage()` instead of forwarding
`classified.Message`.
**Frontend**
- Removed the parallel frontend message-generation system:
`getRetryMessage()`, `getProviderDisplayName()`,
`getRetryProviderSubject()`, and the `PROVIDER_DISPLAY_NAMES` map are
all deleted from `chatStatusHelpers.ts`.
- `liveStatusModel.ts` passes `retryState.error` through directly — the
backend owns the copy.
- Added specific title and retry copy for `startup_timeout`, and
extended the title mapping to cover `auth` and `config`.
- Kind pills now show humanized labels ("Startup timeout", "Rate limit",
etc.) instead of raw enum strings.
- Removed the redundant "Provider anthropic" metadata row.
- Removed the terminal "Retryable" / "Not retryable" badge.
- Normalized `"usage-limit"` → `"usage_limit"` and added it to
`ChatProviderFailureKind` so all error kinds follow the same underscore
convention and live in one enum.
Refs #23282.
## Summary
This change removes the steady-state "resolve the latest workspace
agent" query from chat execution.
Instead of asking the database for the latest build's agent on every
turn, a chat now persists the workspace/build/agent binding it actually
uses and reuses that binding across subsequent turns. The common path
becomes "load the bound agent by ID and dial it", with fallback paths to
repair the binding when it is missing, stale, or intentionally changed.
## What changes
- add `workspace_id`, `build_id`, and `agent_id` binding fields to
`chats`
- expose those fields through the chat API / SDK so the execution
context is explicit
- load the persisted binding first in chatd, instead of always resolving
the latest build's agent
- persist a refreshed binding when chatd has to re-resolve the workspace
agent
- keep child / subagent chats on the same bound workspace context by
inheriting the parent binding
- leave `build_id` / `agent_id` unset for flows like `create_workspace`,
then bind them lazily on the next agent-backed turn
## Runtime behavior
The binding is treated as an optimistic cache of the agent a chat should
use:
- if the bound agent still exists and dials successfully, we use it
without a latest-build lookup
- if the bound agent is missing or no longer reachable, chatd
re-resolves against the latest build and persists the new binding
- if a workspace mutation changes the chat's target workspace, the
binding is updated as part of that mutation
To avoid reintroducing a hot-path query, dialing uses lazy validation:
- start dialing the cached agent immediately
- only validate against the latest build if the dial is still pending
after a short delay
- if validation finds a different agent, cancel the stale dial, switch
to the current agent, and persist the repaired binding
## Result
The hot path stops issuing
`GetWorkspaceAgentsInLatestBuildByWorkspaceID` for every user message,
which is the source of the DB pressure this PR is addressing. At the
same time, chats still converge to the correct workspace agent when the
binding becomes stale due to rebuilds or explicit workspace changes.
Problem: previously, the deployment-wide chat template allowlist was never actually wired in from `chatd.go`
- Extracts `parseChatTemplateAllowlist` into shared `coderd/util/xjson.ParseUUIDList`
- Adds `Server.chatTemplateAllowlist()` method that reads the allowlist from DB
- Passes `AllowedTemplateIDs` callback to `ListTemplates`, `ReadTemplate`, and `CreateWorkspace` tool constructors
> 🤖 Created by Coder Agents and reviewed by a human.
- Clarify the system prompt to prefer tools before asking the user for
clarification.
- Limit clarification to cases where ambiguity or user preferences
materially affect the outcome.
- Remove the contradictory instruction to always start by asking
clarifying questions.
> 🤖 This PR has been reviewed by the author.
## Summary
Adds a general-purpose `map[string]string` label system to chats, stored
as jsonb with a GIN index for efficient containment queries.
This is a standalone foundational feature that will be used by the
upcoming Automations feature for session identity (matching webhook
events to existing chats), replacing the need for bespoke session-key
tables.
## Changes
### Database
- **Migration 000451**: Adds `labels jsonb NOT NULL DEFAULT '{}'` column
to `chats` table with a GIN index (`idx_chats_labels`)
- **`InsertChat`**: Accepts labels on creation via `COALESCE(@labels,
'{}')`
- **`UpdateChatByID`**: Supports partial update —
`COALESCE(sqlc.narg('labels'), labels)` preserves existing labels when
NULL is passed
- **`GetChats`**: New `has_labels` filter using PostgreSQL `@>`
containment operator
- **`GetAuthorizedChats`**: Synced with generated `GetChats` (new column
scan + query param)
### API
- **Create chat** (`POST /chats`): Accepts optional `labels` field,
validated before creation
- **Update chat** (`PATCH /chats/{chat}`): Supports `labels` field for
atomic label replacement
- **List chats** (`GET /chats`): Supports `?label=key:value` query
parameters (multiple are AND-ed)
### SDK
- `Chat`, `CreateChatRequest`, `UpdateChatRequest`, `ListChatsOptions`
all gain `Labels` fields
- `UpdateChatRequest.Labels` is a pointer (`*map[string]string`) so
`nil` means "don't change" vs empty map means "clear all"
### Validation (`coderd/httpapi/labels.go`)
- Max 50 labels per chat
- Key: 1–64 chars, must match `[a-zA-Z0-9][a-zA-Z0-9._/-]*` (supports
namespaced keys like `github.repo`, `automation/pr-number`)
- Value: 1–256 chars
- 13 test cases covering all edge cases
### Chat runtime
- `chatd.CreateOptions` gains `Labels` field, threaded through to
`InsertChat`
- Existing `UpdateChatByID` callers (e.g., quickgen title updates) are
unaffected — NULL labels preserve existing values via COALESCE
We had a bug where computer use base64-encoded screenshots would not be
interpreted as screenshots anymore once saved to the db, loaded back
into memory, and sent to Anthropic. Instead, they would be interpreted
as regular text. Once a computer use agent made enough screenshots and
stopped, and you tried sending it another message, you'd get an out of
context error:
<img width="808" height="367" alt="Screenshot 2026-03-23 at 12 02 54"
src="https://github.com/user-attachments/assets/f0bf6be2-4863-47ca-a7a9-9e6d9dfceeed"
/>
This PR fixes that.
## Summary
Introduces a new `context-file` ChatMessagePart type for persisting
workspace instruction files (AGENTS.md) as durable, frontend-visible
message parts. This is the foundation for showing loaded context files
in the chat input's context indicator tooltip.
### Problem
Previously, instruction files were resolved transiently on every turn
via `resolveInstructions()` → `InsertSystem()` and injected into the
in-memory prompt without persistence. The frontend had no knowledge that
instruction files were loaded into context, and there was no way to
surface this information to users.
### Solution
Instruction files are now read **once** when a workspace is first
attached to a chat (matching how [openai/codex handles
it](https://developers.openai.com/codex/guides/agents-md)) and persisted
as `user`-role, `both`-visibility message parts with a new
`context-file` type. This ensures:
- **Durability**: survives page refresh (data is in the DB, returned by
`getChatMessages`)
- **Cache-friendly**: `user`-role avoids the system-message hoisting
that providers do, keeping the instruction content in a stable position
for prompt caching
- **Frontend-visible**: the frontend receives paths and truncation
status for future context indicator rendering
- **Extensible**: the same pattern works for Skills (future)
### Key changes
| Layer | Change |
|---|---|
| **SDK** (`codersdk/chats.go`) | Add `ChatMessagePartTypeContextFile`
with `context_file_path`, `context_file_content` (internal, stripped
from API), `context_file_truncated` fields |
| **Prompt expansion** (`chatprompt`) | Expand `context-file` parts to
`<workspace-context>` text blocks in `partsToMessageParts()` |
| **Chat engine** (`chatd.go`) | Add `persistInstructionFiles()`, called
on first turn with a workspace. Remove per-turn `resolveInstructions()`
+ `InsertSystem()` from `processChat()` and `ReloadMessages` |
| **Frontend** | Ignore `context-file` parts in `messageParsing.ts` and
`streamState.ts` (no rendering yet — follow-up will add tooltip display)
|
### How it works
1. On each turn, `processChat` checks if any loaded message contains
`context-file` parts
2. If not (first turn with a workspace), reads AGENTS.md files via the
workspace agent connection and persists them
3. For this first turn, also injects the instruction text into the
prompt (since messages were loaded before persistence)
4. On all subsequent turns, `ConvertMessagesWithFiles()` encounters the
persisted `context-file` parts and expands them into text automatically
— no extra resolution needed
The `ComputerUseProviderTool` function needed a little bit of an
adjustment because I changed `NewComputerUseTool`'s signature in
upstream fantasy a little bit.
- Stores a deployment-wide agents template allowlist in `site_configs`
(`agents_template_allowlist`)
- Adds `GET/PUT /api/experimental/chats/config/template-allowlist`
endpoints
- Filters `list_templates`, `read_template`, and `create_workspace` chat
tools by allowlist, if defined (empty=all allowed)
- Add "Templates" admin settings tab in Agents UI ([what it looks
like](https://624de63c6aacee003aa84340-sitjilsyrr.chromatic.com/?path=/story/pages-agentspage-agentsettingspageview--template-allowlist))
> 🤖 This PR was created with the help of Coder Agents, and has been
reviewed by my human. 🧑💻
## Problem
When chatd pushes a branch and then creates a PR (e.g. `git push`
followed by `gh pr create`), the gitsync background worker often picks
up the stale `chat_diff_statuses` row between the two operations. At
that point no PR exists yet, so the worker skips the row. However, the
acquisition SQL locks the row for **5 minutes** (crash-recovery
interval), creating a dead zone where the PR diff is invisible in the UI
until the user manually navigates to the chat.
### Root cause
1. `git push` triggers `GIT_ASKPASS` → coderd external-auth handler →
`MarkStale()` sets `stale_at = now - 1s`
2. Background worker acquires the row within ~10s, atomically bumps
`stale_at = NOW() + 5 min` (crash-recovery lock)
3. Worker calls `ResolveBranchPullRequest` → no PR exists yet → returns
`nil` → worker skips with `continue`
4. `gh pr create` completes moments later, but uses its own auth (not
`GIT_ASKPASS`), so no second `MarkStale` fires
5. Row is locked for 5 minutes before the worker can retry
Loading the chat works immediately because `GET /chats/{chat}` calls
`resolveChatDiffStatus` synchronously, which discovers the PR inline.
## Fix
When `ResolveBranchPullRequest` returns nil (no PR yet) **and** the row
was recently marked stale (within 2 minutes), apply a short 15-second
backoff via `BackoffChatDiffStatus` instead of letting the 5-minute
acquisition lock stand. Outside the retry window, the worker skips the
row as before — no indefinite fast-polling for branches that never
receive a PR.
To make the "recently marked stale" check work, `updated_at` is no
longer overwritten by the acquisition and backoff SQL queries. This
preserves it as a reliable "last externally changed" timestamp (set by
`MarkStale` or a successful refresh).
### Behavior summary
| Scenario | `updated_at` age | Backoff | Effective retry |
|---|---|---|---|
| Fresh push, no PR yet | < 2 min | 15s (`NoPRBackoff`) | ~15s |
| Old row, no PR | ≥ 2 min | None (skip) | ~5 min (acquisition lock) |
| Error (any age) | Any | 120s (`DiffStatusTTL`) | ~120s |
| Success (any age) | Any | 120s (`DiffStatusTTL`) | ~120s |
## Changes
- **`coderd/database/queries/chats.sql`** — Remove `updated_at = NOW()`
from `AcquireStaleChatDiffStatuses` and `BackoffChatDiffStatus`
- **`coderd/database/queries.sql.go`** — Regenerated
- **`coderd/x/gitsync/worker.go`** — Add `NoPRBackoff` (15s) and
`NoPRRetryWindow` (2 min) constants; apply short backoff only within the
retry window
- **`coderd/x/gitsync/worker_test.go`** — Add
`TestWorker_NoPR_RecentMarkStale_BacksOffShort` and
`TestWorker_NoPR_OldRow_Skips`
- Add `SanitizePromptText` stripping ~24 invisible Unicode codepoints
and collapsing excessive newlines
- Apply at write and read paths for defense-in-depth
- Frontend: warn in both prompt textareas when invisible characters
detected
- Explicit codepoint list (not blanket `unicode.Cf`) to avoid breaking
flag emoji
- 34 Go tests + idempotency meta-test, 11 TS unit tests, 4 Storybook
stories
> This PR was created with the help of Coder Agents, and was reviewed by my human.
- Add `X-Coder-Owner-Id`, `X-Coder-Chat-Id`, `X-Coder-Subchat-Id`,
`X-Coder-Workspace-Id` headers to all outgoing LLM API requests from
chatd
- Extend `ModelFromConfig` with `extraHeaders` param, forwarded via
Fantasy `WithHeaders` on all 8 providers
- Add `CoderHeaders(database.Chat)` helper to build the header map from
chat state
- Update all 4 `ModelFromConfig` call sites (resolveChatModel,
computer-use override, title gen, push summary)
- Thread `database.Chat` into `generatePushSummary` (was `chatTitle
string`)
- Tests: `TestCoderHeaders` (4 subtests),
`TestModelFromConfig_ExtraHeaders` (OpenAI + Anthropic),
`TestModelFromConfig_NilExtraHeaders`
- Refactor existing `TestModelFromConfig_UserAgent` to use channel-based
signaling
> 🤖 This PR was generated by Coder Agents and self-reviewed by a human.
create_workspace could create a replacement workspace after a single 5s
agent dial failed, even when the existing workspace agent had recently
checked in. That made temporary reachability blips look like dead
workspaces and let chatd replace a running workspace too aggressively.
Use the workspace agent's DB-backed status with the deployment's
AgentInactiveDisconnectTimeout before allowing replacement. Recently
connected and still-connecting agents now reuse the existing workspace,
while disconnected or timed-out agents still allow a new workspace. This
also threads the inactivity timeout through chatd and adds focused
coverage for the reuse and replacement branches.
## Problem
When an MCP tool returns an `EmbeddedResource` content item (e.g. GitHub
MCP server returning file contents via `get_file_contents`), the
`convertCallResult` function falls through to the `default` case,
producing:
```
[unsupported content type: mcp.EmbeddedResource]
```
This loses the actual resource content and shows an unhelpful message in
the chat UI.
## Root Cause
The type switch in `convertCallResult` handles `TextContent`,
`ImageContent`, and `AudioContent`, but not the other two `mcp.Content`
implementations from `mcp-go`:
- `mcp.EmbeddedResource` — wraps a `ResourceContents` (either
`TextResourceContents` or `BlobResourceContents`)
- `mcp.ResourceLink` — contains a URI, name, and description
## Fix
Add two new cases to the type switch:
1. **`mcp.EmbeddedResource`**: nested type switch on `.Resource`:
- `TextResourceContents` → append `.Text` to `textParts`
- `BlobResourceContents` → base64-decode `.Blob` as binary (type
`"image"` or `"media"` based on MIME)
- Unknown → fallback `[unsupported embedded resource type: ...]`
2. **`mcp.ResourceLink`**: render as `[resource: Name (URI)]` text
## Testing
Added 3 new test cases (all passing, full suite 23/23 PASS):
- `TestConnectAll_EmbeddedResourceText` — text resource extraction
- `TestConnectAll_EmbeddedResourceBlob` — binary blob decoding
- `TestConnectAll_ResourceLink` — resource link rendering
Child chats created via `spawn_agent` and `spawn_computer_use_agent`
were not inheriting the parent's `MCPServerIDs`, meaning subagents lost
access to the parent's MCP server tools.
## Changes
- Pass `parent.MCPServerIDs` in the `CreateOptions` for both
`createChildSubagentChat()` and the `spawn_computer_use_agent` tool
handler in `coderd/x/chatd/subagent.go`.
## Tests
Added 3 tests in `subagent_internal_test.go`:
- `TestCreateChildSubagentChat_InheritsMCPServerIDs` — verifies child
chat gets parent's MCP server IDs (multiple servers)
- `TestSpawnComputerUseAgent_InheritsMCPServerIDs` — verifies computer
use subagent gets parent's MCP server IDs via the tool
- `TestCreateChildSubagentChat_NoMCPServersStaysEmpty` — verifies no
regression when parent has no MCP servers
*Disclaimer: implemented by a Coder Agent and reviewed by me.*
Renames the env vars used by chatd integration tests from the canonical
`SOMEPROVIDER_API_KEY` (e.g. `ANTHROPIC_API_KEY`, `OPENAI_API_KEY`) to
`SOMEPROVIDER_TEST_API_KEY` (e.g. `ANTHROPIC_TEST_API_KEY`,
`OPENAI_TEST_API_KEY`) so that test-specific keys don't collide with
production/canonical provider credentials.
Relates to https://github.com/coder/internal/issues/1425
See also:
https://codercom.slack.com/archives/C0AGTPWLA3U/p1774433646799499
## Summary
Adds `xhigh` to the OpenAI reasoning effort normalizer so GPT-5.4 class
models can use `reasoning_effort: xhigh` without it being silently
dropped.
## Problem
The SDK schema (`codersdk/chats.go`) already advertises `xhigh` as a
valid `reasoning_effort` value, but the runtime normalizer in
`chatprovider.go` only accepts `minimal|low|medium|high` for the OpenAI
provider. When a user sets `xhigh`, `ReasoningEffortFromChat()` returns
`nil` and the value never reaches the OpenAI API.
## Changes
- **Fantasy dependency**: Updated `kylecarbs/fantasy` (cj/go1.25) which
now includes the `ReasoningEffortXHigh` constant
([kylecarbs/fantasy#9](https://github.com/kylecarbs/fantasy/pull/9)).
- **`chatprovider.go`**: Adds `fantasyopenai.ReasoningEffortXHigh` to
the OpenAI case in `ReasoningEffortFromChat()`.
- **`chatprovider_test.go`**: Adds `OpenAIXHighEffort` test case.
## Upstream
-
[charmbracelet/fantasy#186](https://github.com/charmbracelet/fantasy/pull/186)
> **PR Stack**
> 1. #23351 ← `#23282`
> 2. #23282 ← `#23275`
> 3. **#23275** ← `#23349` *(you are here)*
> 4. #23349 ← `main`
---
## Summary
Extracts a structured error classification subsystem for agent chat
(`chatd`) so that retry and error payloads carry machine-readable
metadata — error kind, provider name, HTTP status code, and retryability
— instead of raw error strings.
This is the **backend half** of the error-handling work. The frontend
counterpart is in #23282.
## Changes
### New package: `coderd/chatd/chaterror/`
Canonical error classification — extracts error kind, provider, status
code, and user-facing message from raw provider errors. One source of
truth that drives both retry policy and stream payloads.
- **`kind.go`**: Error kind enum (`rate_limit`, `timeout`, `auth`,
`config`, `overloaded`, `unknown`).
- **`signals.go`**: Signal extraction — parses provider name, HTTP
status code, and retryability from error strings and wrapped types.
- **`classify.go`**: Classification logic — maps extracted signals to an
error kind.
- **`message.go`**: User-facing message templates keyed by kind +
signals.
- **`payload.go`**: Projectors that build `ChatStreamError` and
`ChatStreamRetry` payloads from a classified error.
### Modified
- **`codersdk/chats.go`**: Added `Kind`, `Provider`, `Retryable`,
`StatusCode` fields to `ChatStreamError` and `ChatStreamRetry`.
- **`coderd/chatd/chatretry/`**: Thinned to retry-policy only;
classification logic moved to `chaterror`.
- **`coderd/chatd/chatloop/`**: Added per-attempt first-chunk timeout
(60 s) via `guardedStream` wrapper — produces retryable
`startup_timeout` errors instead of hanging forever.
- **`coderd/chatd/chatd.go`**: Publishes normalized retry/error payloads
via `chaterror` projectors.
Nine subtests covering the poll loop, pubsub notification path,
timeout, context cancellation, descendant auth check, and both
error-status branches in handleSubagentDone.
Wire p.clock through awaitSubagentCompletion's timer and ticker
so future tests can use quartz mock clock. Tests use channel-based
coordination and context.WithTimeout instead of time.Sleep.
Coverage: awaitSubagentCompletion 0%->70.3%, handleSubagentDone
0%->100%, checkSubagentCompletion 0%->77.8%,
latestSubagentAssistantMessage 0%->78.9%.
## Summary
Large pasted text that the UI collapses into an attachment chip was
completely invisible to the LLM. Providers only accept specific MIME
types (images, PDFs) in file content blocks — a `text/plain` `FilePart`
is silently dropped, so the model received nothing for pasted content.
## Fix
Detect paste-originated text files by their
`pasted-text-{timestamp}.txt` filename pattern and convert them to
`fantasy.TextPart` with a bounded 128 KiB inline body and truncation
notice. Binary uploads and real uploaded text files keep their existing
`FilePart` semantics.
The detection uses the existing frontend naming convention
(`pasted-text-YYYY-MM-DD-HH-MM-SS.txt`) combined with a text-like MIME
check for defense-in-depth. A TODO marks this for migration to explicit
origin metadata.
<details>
<summary>Review notes: intentionally skipped findings</summary>
A 10-reviewer deep review was run on this change. The following findings
were raised and intentionally dropped after cross-check. Documenting
them here so future reviewers do not re-flag the same concerns:
**"Unresolved file IDs cause silent data loss" (Edge Case Analyst P1)**
— When a file ID is not in the resolver map, `name` stays empty and
paste detection fails. This is pre-existing behavior for ALL file types
(not introduced by this change). The resolver calls `GetChatFilesByIDs`
which returns whatever rows exist; missing IDs simply fall through to an
empty `FilePart`. The Contract Auditor independently traced this path
and confirmed the fallback is safe. If the file was deleted between
message construction and conversion, the model already saw nothing
before this patch — this change does not make it worse.
**"String builder pre-allocation overhead" (Performance Analyst P1)** —
Misidentified scope. `formatSyntheticPasteText` is only called when
`isSyntheticPaste` returns true (actual synthetic pastes), not for every
file part. The `Grow()` call is correct and efficient.
**"Constant naming violates Uber style" (Style Reviewer P1)** —
Over-severity. `syntheticPasteInlineBudget` is standard Go camelCase for
unexported constants, consistent with the Uber guide and surrounding
code.
**"`IsSyntheticPasteForTest` naming is misleading" (Style Reviewer P2)**
— This is the standard Go `export_test.go` pattern. The `ForTest` suffix
is conventional.
</details>
Linear's MCP server (`mcp.linear.app`) returns `token_type="bearer"`
(lowercase) in its OAuth2 token response but rejects requests that use
the lowercase form in the `Authorization` header. RFC 6750 says the
scheme is case-insensitive, but Linear enforces capital-B `Bearer`.
Confirmed by running the actual Linear MCP OAuth flow end-to-end:
- `Authorization: Bearer <token>` → **42 tools, works**
- `Authorization: bearer <token>` → **401 invalid_token**
This is a one-line fix: normalize any case variant of `bearer` to
`Bearer` before building the `Authorization` header, matching the
behavior of the mcp-go library's own OAuth handler.
Adds a `propose_plan` tool that presents a workspace markdown file as a
dedicated plan card in the agent UI.
The workflow is: the agent uses `write_file`/`edit_files` to build a
plan file (e.g. `/home/coder/PLAN.md`), then calls `propose_plan(path)`
to present it. The backend reads the file via `ReadFile` and the
frontend renders it as an expanded markdown preview card.
**Backend** (`coderd/x/chatd/chattool/proposeplan.go`): new tool
registered as root-chat-only. Validates `.md` suffix, requires an
absolute path, reads raw file content from the workspace agent. Includes
1 MiB size cap.
**Frontend** (`site/src/components/ai-elements/tool/`): dedicated
`ProposePlanTool` component with `ToolCollapsible` + `ScrollArea` +
`Response` markdown renderer, expanded by default. Custom icon
(`ClipboardListIcon`) and filename-based label.
**System prompt** (`coderd/x/chatd/prompt.go`): added `<planning>`
section guiding the agent to research → write plan file → iterate → call
`propose_plan`.
Kyle Carberry