Extend the delete_deleted_user_resources() trigger so that secrets
belonging to a soft-deleted user are removed in the same transaction as
the existing api_keys and user_links cleanup.
user_secrets.user_id has ON DELETE CASCADE, but Coder soft-deletes users
by flipping users.deleted rather than removing the row, so the foreign key
cascade never fires and secrets would otherwise survive deletion.
Assisted by Coder Agents.
Use typed atomics (atomic.Int64, atomic.Int32, etc.) in test files to prevent
mixing atomic and non-atomic access on the same value, guarantee 64-bit
alignment on 32-bit platforms, and provide a cleaner API.
## Summary
- Adds the `coder-agents-review` agent skill that drives PRs through the
`coder-agents-review` GitHub app review loop
- The skill automates: triggering review, waiting for feedback, fixing
issues, validating, and re-requesting until approved
## Test plan
- [ ] Verify the skill file is well-formed and the loop instructions are
correct
- [ ] Test triggering the skill on a PR with `/coder-agents-review`
Made with [Cursor](https://cursor.com)
---------
Co-authored-by: Cursor <cursoragent@cursor.com>
Added a coder-agents-review skill covering:
- when to request review
- how to paginate app activity before deriving state
- how to require exact app identity and explicit approval evidence
> Mux is acting on Mike's behalf.
The first `/mcp/tools` request could race workspace startup and return
an empty tool list before startup scripts had a chance to write
`.mcp.json`. Chatd may only discover tools once for a turn, so that
empty response could hide workspace MCP tools even though the agent
loaded them later.
Make the manager wait for startup to settle before treating missing MCP
config files as a real empty state. Tool listing now goes through one
manager-owned path that starts reload work independently of caller
cancellation; caller contexts only bound that caller's wait. After the
first reload body settles, transient reload errors return cached tools
with the error so the HTTP handler can degrade to the last known tool
set instead of returning `[]`.
The handler is intentionally thin: it asks the manager for tools, logs
any degraded path, and still returns the tool response shape callers
already expect. Tests cover startup gating, caller-canceled waits,
manager close, reload timeout via quartz, and cached-tool fallback after
a later reload error.
TestPromoteQueuedWhileRequiresActionMixedTools has flaked three times across
Windows and Ubuntu CI runners since 2026-05-06; local repro on the dev
workspace has not surfaced it. The May 8 Ubuntu log shows all four
PromoteQueued post-TX pubsub publishes reaching pg_notify, yet the test still
times out 25s later, so the failure is downstream between the subscriber's
listener and the test's events channel. Adds three Debug-level markers in
chatd.go (no logic change) plus two t.Logf markers in the test's reader so
the next CI occurrence pins down exactly which step failed.
Closes ENG-2645
Closescoder/internal#1523
Parallel subtests in `coderd/x/chatd` reused a parent test context with
a `testutil.WaitLong` deadline, so the context could expire before a
subtest was scheduled under load. That made the subagent lifecycle tools
return plain-text context errors instead of the expected JSON payload,
causing flaky JSON unmarshal failures.
Create fresh `chatdTestContext` values inside the affected parallel
subtests and add `chatdTestContext` to the `paralleltestctx` custom
function list so this pattern is caught by `make lint`.
Closes https://github.com/coder/internal/issues/1494
## Summary
Adds a `stop_workspace` tool to chatd so the model can recover from the
"workspace running but agent dead" failure mode (e.g. an OOM that leaves
the workspace running but the agent unreachable) by stopping and then
starting the workspace.
<img width="924" height="742" alt="image"
src="https://github.com/user-attachments/assets/279dedb6-6e29-4fe1-8754-3a1f01e538bf"
/>
## What changed
**New `stop_workspace` chatd tool**
(`coderd/x/chatd/chattool/stopworkspace.go`). Mirrors `start_workspace`:
shares `WorkspaceMu` to serialize with create/start, waits for any
in-progress build before issuing a stop, and is idempotent only after a
successful Stop transition. Failed stop builds re-attempt rather than
reporting success.
**New `chatStopWorkspace` coderd hook** (`coderd/exp_chats.go`). Mirrors
`chatStartWorkspace` minus the `RequireActiveVersion` gate. Stop should
not be blocked by template version policy.
**Differentiated recovery sentinels** (`coderd/x/chatd/chatd.go`).
`errChatAgentDisconnected` instructs the model to call `stop_workspace`
then `start_workspace`. `errChatDialTimeout` instructs a single retry,
then user escalation if it repeats. The previous single message
conflated transient and persistent failures.
**Two-signal recovery gate.** Recovery is only surfaced when a tool call
times out *and* a fresh DB read of the latest workspace agent says
`Disconnected`. The previous draft escalated on the DB read alone, which
would fire on a 30-second heartbeat blip (e.g. agent respawn) and prompt
a destructive stop/start unnecessarily.
**Cache-hit disconnected handling** now clears the cache and retries a
fresh dial before escalating, rather than returning the recovery
sentinel immediately. Latest-agent classification uses
`GetWorkspaceAgentsInLatestBuildByWorkspaceID` instead of the chat's
bound `AgentID`, so stale bindings after a rebuild don't misclassify.
**Shared chattool helpers** in `coderd/x/chatd/chattool/chattool.go`:
`latestWorkspaceBuildAndJob`, `publishBuildBinding`,
`provisionerJobTerminal`. Applied to both `start_workspace` and
`stop_workspace`.
## Notes
- Reverts an earlier draft that widened `ask_user_question` to root
standard turns. Plan-mode-only behavior is restored.
- The `stop_workspace` tool currently renders via the generic chat
tool-call UI. A follow-up frontend PR will prettify the `stop_workspace`
tool and style it like the `start_workspace` tool.
- Never-connected (`Timeout` status) agents are intentionally excluded
from recovery. They indicate template or startup failure, not the
running-but-dead case this PR targets.
Closes CODAGT-315
#11918 took away advanced settings during template creation however it
did not clean up the documentation of a reference to customising the
template permissions during template creation -
https://coder.com/docs/admin/templates/template-permissions
> By default the Everyone group is assigned to each template meaning any
Coder user can use the template to create a workspace. To prevent this,
disable the Allow everyone to use the template setting when creating a
template.
This setting is no longer present in Coder, so removing it from the
docs.
Closes
[CODAGT-317](https://linear.app/codercom/issue/CODAGT-317/pr-workspaces-sometimes-require-name-confirmation-to-delete).
## Problem
The `/agents` archive-and-delete molly-guard (typing the workspace name)
was firing for chats that had clearly created their own workspace. The
heuristic in `resolveArchiveAndDeleteAction` decides whether
confirmation is needed by comparing the workspace's `created_at` against
the chat's `created_at`:
```ts
return new Date(workspaceCreatedAt) >= new Date(chatCreatedAt);
```
That assumption breaks for **prebuilt workspaces**.
`ClaimPrebuiltWorkspace` rewrites `owner_id`, `name`, `updated_at`,
`last_used_at`, etc., but **never touches `created_at`**, which still
reflects when the prebuild was provisioned by the reconciler, often
hours before the chat exists. Result: every prebuild-claimed workspace
looks pre-existing, so the molly-guard fires.
Concrete example from a real chat:
| Field | Value |
|---|---|
| `chat.created_at` | `2026-05-07T15:12:23Z` |
| `workspace.created_at` (provision) | `2026-05-07T14:22:24Z` |
| `latest_build.created_at` (claim) | `2026-05-07T15:19:09Z` |
`14:22:24 < 15:12:23` so `isWorkspaceAutoCreated` returned false even
though the chat issued the claim.
## Fix (frontend-only)
Derive the moment a workspace was acquired from existing build history
rather than relying on `workspace.created_at`:
- Build #1 initiator = prebuilds system user → workspace was a prebuild
→ use `build_2.created_at` (the claim build) as the acquisition time.
- Build #1 initiator = real user → workspace was created from scratch →
use `workspace.created_at` (unchanged behavior).
- Unclaimed prebuild or no build history → return `null` (force
confirmation; safe degradation for a destructive flow).
The resolver fetches the build list via the existing
`getWorkspaceBuilds` endpoint when the dialog might fire. No new column,
no migration, no schema change. Works retroactively for all existing
claimed prebuilds; no backfill needed.
The prebuilds system user UUID is exposed via
`codersdk.PrebuildsSystemUserID` and typegen'd to `typesGenerated.ts`.
`coderd/database.PrebuildsSystemUserID` parses that constant via
`uuid.MustParse` so the two cannot drift; if the codersdk literal ever
changes, package init fails fast.
## History
The first draft of this PR added a `workspaces.claimed_at` column
populated by `ClaimPrebuiltWorkspace`. After review feedback from
@johnstcn pointing out that the same fact is already implicit in build
history, I pivoted to the frontend-only approach. Subsequent review
notes consolidated the prebuilds system user UUID into a single
typegen'd constant.
## Why not the other open PRs
- **#25055** (`chatKey` cache fallback) only fixes a different
cache-miss path; it explicitly notes it does not address `created_at <
chat.created_at`.
- **#25053** (`chats.workspace_auto_created` boolean) puts the truth on
the wrong side of the schema: "this workspace was claimed at time T" is
a property of the workspace, not the chat. The MCP plumbing it adds is
also unnecessary now that the same answer is available from build
history.
## Test plan
- `pnpm vitest run --project=unit
src/pages/AgentsPage/utils/agentWorkspaceUtils.test.ts` — 40/40 pass;
new cases cover prebuild claim before/after chat, unclaimed prebuild,
missing-build-history fallback, and the fetch-skip when the chat is not
in cache.
- `pnpm lint:types`, `pnpm check`, `make pre-commit`.
<details>
<summary>Disclosure</summary>
Opened on behalf of @kylecarbs by [Coder
Agents](https://coder.com/coder-agents).
</details>
# Summary
Implements
https://linear.app/codercom/issue/AIGOV-282/add-ai-model-price-table-and-seed-generator
This PR lays the groundwork for AI Bridge cost controls (per the AI
Governance RFC). It adds the foundation needed for future cost tracking:
a place to store per-model token prices, a way to keep those prices in
sync with upstream pricing data, and a startup mechanism that ensures
every deployment has prices loaded before AI Bridge starts processing
requests.
The price data comes from [models.dev](https://models.dev/), a
community-maintained catalogue of AI provider pricing. A generator
script fetches the latest prices, filters to Anthropic and OpenAI for
now, and produces a seed file checked into the repository.
On every server startup the seed is applied to the database, so new
releases automatically pick up any price corrections that landed since
the previous one. Existing rows are overwritten with the latest prices;
rows for models no longer in the seed are left untouched.
# Batching the AI model price seed: three approaches
Context: at server startup we seed the `ai_model_prices` table from an
embedded JSON price book (~70 rows today, will grow as we add providers,
potentially 4000+).
Each row is:
```text
(provider, model, input_price, output_price, cache_read_price, cache_write_price)
```
Any of the four price columns can be:
- `NULL` → “price unknown for this dimension”
- explicit `0` → “free”
The batch must be an UPSERT so re-running is idempotent and existing
rows pick up new prices.
We considered three implementations.
---
## Approach 1 — Per-row UPSERT in a Go loop
```go
for _, row := range rows {
if err := db.UpsertAIModelPrice(ctx, database.UpsertAIModelPriceParams{
Provider: row.Provider,
Model: row.Model,
InputPrice: nullInt64(row.InputPrice),
// ...
}); err != nil {
return err
}
}
```
### Pros
- Trivial.
- NULL handling falls out naturally from `sql.NullInt64`.
### Cons
- `N` round-trips per seed.
- With ~70 rows that means ~70 statement executions on every startup,
even inside a transaction.
- Doesn't scale gracefully as the price book grows, potentially 4000+.
---
## Approach 2 — `UNNEST` with parallel arrays
Pass each column as a separate Go slice. Postgres unnests them in
parallel into a virtual table, then `INSERT ... SELECT`.
```sql
INSERT INTO ai_model_prices (
provider,
model,
input_price,
output_price,
cache_read_price,
cache_write_price
)
SELECT
UNNEST(@providers::text[]),
UNNEST(@models::text[]),
NULLIF(UNNEST(@input_prices::bigint[]), -1),
NULLIF(UNNEST(@output_prices::bigint[]), -1),
NULLIF(UNNEST(@cache_read_prices::bigint[]), -1),
NULLIF(UNNEST(@cache_write_prices::bigint[]), -1)
ON CONFLICT (provider, model) DO UPDATE SET
input_price = EXCLUDED.input_price,
output_price = EXCLUDED.output_price,
cache_read_price = EXCLUDED.cache_read_price,
cache_write_price = EXCLUDED.cache_write_price,
updated_at = NOW();
```
Go side: flatten rows into six parallel slices.
Use a sentinel (`-1`) for “missing”, since `lib/pq` can't encode `NULL`
into a `bigint[]` element.
```go
providers := make([]string, len(rows))
models := make([]string, len(rows))
inputs := make([]int64, len(rows))
outputs := make([]int64, len(rows))
cacheR := make([]int64, len(rows))
cacheW := make([]int64, len(rows))
for i, r := range rows {
providers[i] = r.Provider
models[i] = r.Model
inputs[i] = -1
if r.InputPrice != nil {
inputs[i] = *r.InputPrice
}
outputs[i] = -1
if r.OutputPrice != nil {
outputs[i] = *r.OutputPrice
}
cacheR[i] = -1
if r.CacheReadPrice != nil {
cacheR[i] = *r.CacheReadPrice
}
cacheW[i] = -1
if r.CacheWritePrice != nil {
cacheW[i] = *r.CacheWritePrice
}
}
return db.UpsertAIModelPrices(ctx, database.UpsertAIModelPricesParams{
Providers: providers,
Models: models,
InputPrices: inputs,
OutputPrices: outputs,
CacheReadPrices: cacheR,
CacheWritePrices: cacheW,
})
```
### Pros
- Single round-trip.
### Cons
- The generated `sqlc` params become plain `[]int64`, which can't
represent `NULL`.
---
## Approach 3 — `jsonb_array_elements` over a single `@seed::jsonb`
(chosen)
Pass the raw seed JSON as one parameter; let Postgres expand and parse
it.
```sql
INSERT INTO ai_model_prices (
provider,
model,
input_price,
output_price,
cache_read_price,
cache_write_price
)
SELECT
elem->>'provider',
elem->>'model',
(elem->>'input_price')::bigint,
(elem->>'output_price')::bigint,
(elem->>'cache_read_price')::bigint,
(elem->>'cache_write_price')::bigint
FROM jsonb_array_elements(@seed::jsonb) AS elem
ON CONFLICT (provider, model) DO UPDATE SET
input_price = EXCLUDED.input_price,
output_price = EXCLUDED.output_price,
cache_read_price = EXCLUDED.cache_read_price,
cache_write_price = EXCLUDED.cache_write_price,
updated_at = NOW();
```
Go side reduces to:
```go
return db.UpsertAIModelPrices(ctx, seedJSON)
```
### Pros
- Single round-trip.
- NULLs fall out naturally:
- `(elem->>'cache_write_price')::bigint` becomes `NULL`
- no sentinels
- The seed is already JSON:
- Existing precedent:
- `jsonb_array_elements` is already used elsewhere in the codebase
### Cons
- Less type-safe at the SQL boundary than `UNNEST`
- Slightly less standard than `UNNEST`
- Readers need familiarity with:
- `jsonb_array_elements`
- `->>` extraction syntax
- Postgres pays JSON parse cost
- negligible at our scale
---
---
# Decision
We picked Approach 3.
It collapses the round-trips like `UNNEST` does, but without:
- nullable-array workarounds
- sentinel values
The 5s timeout cancelled cold-start ListMCPTools calls before the
agent's 30s connectTimeout could settle, so workspace MCP tools
never reached the LLM. Bump to 35s and scope to ListMCPTools only.
The badge next to the loading spinner in the agent logs section was
showing `agent.log_sources.length` (total log sources registered on the
agent). This is a static count unrelated to what's actively running.
Now it shows the count of startup scripts still in progress: scripts
where `run_on_start` is true and `status` is not yet set. Scripts
without a `status` haven't completed; completed scripts receive `"ok"`,
`"exit_failure"`, `"timed_out"`, or `"pipes_left_open"`. The badge also
hides when zero scripts are running.
> [!NOTE]
> Generated by Coder Agents
coder/fantasy now fails closed when Anthropic or OpenAI Responses
streams close before their provider terminal events instead of yielding
a successful finish.
This bumps the fantasy replacement to coder/fantasy#33 and teaches chat
error classification to treat those failures as retryable timeout errors
with explicit stream-closed messages.
<img width="875" height="311" alt="image"
src="https://github.com/user-attachments/assets/69c6f7b5-c885-46d2-a88b-b7a2b111bd55"
/>
## Summary
Make Coder's chat agent honest about workspaces that use
`coder_external_agent`. Three behaviors change so the chat stops
pretending it can drive an external workspace through to a usable state
on its own.
<img width="859" height="537" alt="image"
src="https://github.com/user-attachments/assets/0561442b-95f1-4a2d-853c-7e3776114680"
/>
## Problem
External agents are not started by Coder. The user has to run `coder
agent` on their own host with a token Coder generates. Before this
change, the chat agent treated those workspaces like any other:
- `create_workspace` would enqueue a build for an external-agent
template and then wait minutes (~22 worst case) for an agent that was
never going to come up.
- When mid-turn tool calls dialed an external agent that was not
connected, the chat burned the full 30-second dial timeout and returned
generic "the workspace may need to be restarted from the Coder
dashboard" guidance, which is not the action the user can take.
- Nothing told the chat (or the user, through the chat) that the next
action lives outside Coder.
## Fix
Three changes scoped to `coderd/x/chatd/`:
1. **`create_workspace` blocks templates with external agents.** The
tool reads `template_versions.has_external_agent` for the template's
active version and refuses external-agent templates with a message
instructing the chat to pick a different template, or to have the user
create and start the workspace themselves and then attach it.
2. **Attaching an existing external workspace stays open.** No
selection-time gate on attachment; users can still bind a working
external workspace to a chat.
3. **External-agent-aware error handling on connection.** Two
complementary changes both predicated on proven connectivity failures
rather than every dial error:
- **`getWorkspaceConn` preflight and timeout handling.** Before opening
a connection, the cache-miss path reads the agent's status from the
already-loaded row. If the selected agent is external and clearly
offline according to the existing `isAgentUnreachable` helper
(`Disconnected` or `Timeout`, never `Connecting`), it returns an
external-agent-specific error immediately instead of waiting out the
30-second dial timeout. `Connecting` external agents fall through to the
dial so a user who just started the agent on their host can still
succeed in the same turn. The preflight only fires when the agent is
still the latest selected agent for the workspace, so stale-binding
recovery via `dialWithLazyValidation` is unaffected. The post-dial
rewrite is limited to the dial timeout sentinel; stale/no-agent bindings
and non-timeout dial failures preserve their original errors.
- **`waitForAgentReady` timeout-branch rewrite.** The 2-minute retry
loop used by `create_workspace` and `start_workspace` runs unchanged for
all agents. When the loop's outer deadline elapses, the timeout branch
substitutes the external-agent message in place of the raw dial error if
the agent belongs to an external resource.
This applies the same pattern that the cache-hit path of
`getWorkspaceConn` already used (`isAgentUnreachable` returning
`errChatAgentDisconnected`), extended to the cache-miss path and to the
readiness helper, with the external-agent-aware error rewrite layered
only on confirmed offline or timeout paths.
Closes CODAGT-314
Create and start workspace tool cards now recognize `INSUFFICIENT_QUOTA`
results and use the server-provided quota failure title in the build-log
dropdown. The existing warning icon and tooltip remain, while the
assistant response remains the place for the detailed recovery guidance.
Adds Storybook coverage for quota-reached create and start workspace
results.
https://github.com/coder/coder/pull/24956 added the necessary backend
changes.
<img width="897" height="505" alt="image"
src="https://github.com/user-attachments/assets/6cab8798-393d-429f-a3c3-a8ed50402d42"
/>
Closes CODAGT-20
Workspace-agent logs emitted while serving chatd-driven requests were
not correlated with the originating chat, making agent logs hard to
attribute to the corresponding/originating chat.
This adds agent-side chat context middleware that parses `Coder-Chat-Id`
once, enriches agent access logs and structured handler/background logs,
and adds a chatd bridge log when chat headers are attached to an agent
connection.
Closes CODAGT-324
## Summary
Replace the "Premium" label with "AI Governance Add-On" and add a
disclaimer that the AI Governance Add-On is required for AI Gateway and
Agent Firewall as of Coder v2.32, across all AI Governance doc pages and
their children.
## Changes
**Label and requirement updates (7 files):**
- `docs/ai-coder/ai-governance.md`: Removed "(Premium)" from title;
updated GA section to state add-on required as of v2.32.
- `docs/ai-coder/ai-gateway/setup.md`: "Premium license" → "AI
Governance Add-On license".
- `docs/ai-coder/ai-gateway/ai-gateway-proxy/setup.md`: "Premium
license" → "AI Governance Add-On".
- `docs/ai-coder/ai-gateway/clients/claude-code.md`: "(Premium feature)"
→ "(AI Governance Add-On)".
- `docs/manifest.json`: `"state": ["premium"]` → `"state": ["ai
governance add-on"]` for 4 nav entries.
**Disclaimer added to all child pages (26 files):**
AI Gateway pages (18):
`index.md`, `setup.md`, `audit.md`, `monitoring.md`, `mcp.md`,
`reference.md`, `ai-gateway-proxy/index.md`,
`ai-gateway-proxy/setup.md`, `clients/index.md`,
`clients/claude-code.md`, `clients/codex.md`, `clients/mux.md`,
`clients/opencode.md`, `clients/factory.md`, `clients/cline.md`,
`clients/kilo-code.md`, `clients/roo-code.md`, `clients/vscode.md`,
`clients/jetbrains.md`, `clients/zed.md`, `clients/copilot.md`
Agent Firewall pages (8):
`index.md`, `version.md`, `landjail.md`, `rules-engine.md`,
`nsjail/index.md`, `nsjail/docker.md`, `nsjail/k8s.md`, `nsjail/ecs.md`
Other: `security.md`
> [!NOTE]
> The `"ai governance add-on"` state value in `manifest.json` is new.
The docs site renderer may need to be updated to support this state
value.
> Generated by Coder Agents
Fixes
[CODAGT-319](https://linear.app/codercom/issue/CODAGT-319/support-prompt-history-cycling-with-up-arrow).
Pressing the up-arrow key in the agent chat composer now cycles through
prior user prompts in the chat (terminal/Discord/iTerm2 style).
Down-arrow steps forward, Escape exits cycling and restores the
in-progress draft. Cycling is non-destructive: the per-message hover
**Edit** button is still the destructive truncate-and-edit path.
Replaces the previous up-arrow shortcut that immediately entered
destructive history-edit mode (and which had a regression where the
composer rendered as "editing" with an empty input box).
## Behaviour
- **Up** when composer is empty: snapshot the (empty) draft and load the
most recent user prompt; subsequent **Up** presses load older prompts.
Clamp at oldest, no wrap.
- **Up** while non-empty and not yet cycling: pass through (caret
movement preserved).
- Once cycling, **Up / Down** are intercepted unconditionally because
the cycle text fully replaces editor contents. Exit explicitly via
Escape, by sending, or by typing.
- **Down** while cycling: load the next-newer prompt, or restore the
saved draft when past newest.
- **Escape** while cycling: exit cycle and restore the saved draft. This
also applies during streaming; the same keypress is stopped before it
reaches the interrupt handler, and a second Escape interrupts as before.
- **Typing / paste / drop / attach / send / `remountKey` change**: exit
cycle mode and clear the snapshot.
- Cycling is suppressed while `isEditingHistoryMessage`,
`editingQueuedMessageID !== null`, or the input is `disabled` /
`isLoading`.
- Empty `userPromptHistory` makes Up a no-op (no destructive fallback).
## Out of scope (filed as follow-ups if needed)
- Restoring file-reference chips / attachments on cycled messages — v1
cycles plain text only, matching the existing per-message destructive
Edit's `text` payload.
- `^N` / `^P` keybindings (per Cian's note in the Linear thread).
- Per-user "enable/disable history cycling" preference (per Rowan's
note).
- Cross-chat history; cycling is per-chat.
## Tests
New Storybook play functions in `AgentChatInput.stories.tsx`:
- `PromptHistoryCycling` — Up cycles older, clamps at oldest; Down
returns to newer / draft; Escape restores draft.
- `PromptHistoryCyclingExitsOnTyping` — typing exits cycle mode;
subsequent Up snapshots the fresh empty draft and Down restores it.
- `NoPromptHistoryUpArrowIsNoOp` — empty history → Up is a no-op.
- `PromptHistorySuppressedWhileEditingHistoryMessage` — cycling does not
engage while history-editing.
- `PromptHistorySuppressedWhileDisabled` — cycling does not engage while
disabled.
- `PromptHistorySuppressedWhileLoading` — cycling does not engage while
loading.
## Implementation notes
Also rewrites `useImperativeHandle` to delegate to `internalRef.current`
lazily on every call instead of capturing it eagerly at factory time.
The old code crashed when methods were called after a remount because
the captured ref was stale; the new wrapper sees the current Lexical
instance. Behavior changes from throw-on-null to silent no-op, which
matches every other consumer of `ChatMessageInputRef`.
Verified locally:
```
pnpm format
pnpm check
pnpm test:storybook src/pages/AgentsPage/components/AgentChatInput.stories.tsx # 41 passed
pnpm lint
```
## Manual UAT
A 13-case manual UAT covering cycle entry/exit, clamping, draft
restoration, no-history no-op, suppression while editing a history
message, and the send-button enable state — all PASS. Spec lives at the
deleted artifact branch; happy to re-attach if reviewers want it.
<details>
<summary>Implementation plan and decision log</summary>
The complete plan that drove this PR, including design alternatives
considered and edge cases:
```md
# CODAGT-319 — Up-arrow prompt history cycling
## Goal
Pressing the up-arrow key in the agent chat composer should cycle through the
user's previously-sent prompts in the current chat, terminal/Discord/iTerm2
style. Down-arrow steps forward; Escape exits cycle mode and restores the
in-progress draft. Cycling is non-destructive — it only populates the
composer with text the user can choose to resend, edit, or discard.
## Today's behaviour (and the regression)
- `ChatMessageInput` is a Lexical-based plain-text editor used inside `AgentChatInput.tsx`.
- `AgentChatInput.tsx` already wires an `ArrowUp` handler. When the composer is empty and not already editing, it calls `onEditLastUserMessage`.
- `onEditLastUserMessage` puts the user into a destructive "edit history" mode that warns "Editing will delete all subsequent messages and restart the conversation here.".
- Danielle's regression report ("shows me as editing but the input box is empty") indicates the destructive flow has a bug in addition to being the wrong UX for the request. We're replacing that path on the up-arrow, not patching it. The destructive edit remains accessible via the per-message hover Edit button.
## Design
### Behaviour
- Up when composer is empty: snapshot the (empty) draft and load the most recent user prompt. Subsequent Up loads older prompts, clamping at oldest. No wrap.
- Up while non-empty and not yet cycling: pass through. Matches existing gating.
- Once cycling, Up/Down are intercepted unconditionally. Exit via Escape, send, or typing.
- Down while cycling: next-newer or restore draft past newest.
- Escape while cycling: restore draft. During streaming, stop propagation so the same keypress does not interrupt; a second Escape interrupts as before.
- Typing/paste/drop/attach/send: exit cycle.
- Suppressed while isEditingHistoryMessage, editingQueuedMessageID !== null, or disabled/isLoading.
- No history => Up is a no-op.
### State
Local to `AgentChatInput.tsx`:
- `cycleIndex: number | null` — null means not cycling. 0 = newest user prompt.
- `cycleSavedDraft: string | null` — text restored on dismiss.
No localStorage persistence — refresh is a clean exit signal and the chat already has history server-side.
### Wiring
- New prop `userPromptHistory: readonly string[]` on `AgentChatInput`, newest-first.
- Removed `onEditLastUserMessage` prop entirely (its single call-site is being replaced). Removed dead `onEditUserMessage` prop on `ChatPageInput` (no longer needed since the destructive last-message shortcut is gone; the destructive Edit button uses a separate prop chain through `ChatPageTimeline`).
- `ChatPageContent.tsx` derives `userPromptHistory` from existing message store, filtered to `role === "user"` with non-empty `getEditableUserMessagePayload(message).text.trim()`.
### Reset triggers
`cycleIndex` and `cycleSavedDraft` reset on:
1. New `remountKey` (chat change, edit start/cancel).
2. Successful send.
3. Paste, drop, file attach.
4. User typing (detected via `handleContentChange` by comparing the incoming content to `currentCycleValueRef`, the last value applied programmatically).
### Out of scope
- Chip/attachment cycling.
- ^N/^P (Cian's note).
- Per-user toggle (Rowan's note).
- Cross-chat history.
```
</details>
---
> [!NOTE]
> This PR was created on behalf of @ibetitsmike by Coder Agents.
---------
Co-authored-by: Coder Agents <noreply@coder.com>
Adds an `[!IMPORTANT]` callout under the SCIM heading in the OIDC auth
docs noting that Coder's SCIM 2.0 implementation is not a fully
certified or guaranteed implementation of the spec. It covers common
provisioning/deprovisioning flows with major IdPs (Okta, Entra ID, etc.)
but specific attributes, endpoints, or behaviors may not be supported
and may change between releases.
This matches what we say in conversations with prospects and avoids
setting an expectation we can't always meet. Background: #15830 (current
implementation is an MVP scoped to Okta cloud; `PATCH` is not RFC 7644
compliant; user updates only change status, not groups/orgs/roles).
Companion PR: coder/coder.com#738 removes the SCIM row from the pricing
comparison.
> Generated with [Coder Agents](https://coder.com/agents)
Closes#24091
Adds
`TestDeleteChatDebugDataAfterMessageIDStepLevelFieldBoundariesAndNulls`,
which complements the existing triggered-runs test for
`DeleteChatDebugDataAfterMessageID` with boundary and NULL coverage for
step-level message IDs.
The existing
`TestDeleteChatDebugDataAfterMessageIDIncludesTriggeredRuns` already
exercises the `step.assistant_message_id > @message_id` deletion path.
This test focuses on:
- Strict greater-than behavior at the cutoff for assistant and
history-tip step message IDs.
- Step-level assistant and history-tip message ID combinations.
- SQL NULL behavior for step-level message IDs.
- A mixed-step run where one matching step deletes the whole run and
cascades every step.
| Scenario | assistant_message_id | history_tip_message_id | Expected |
|----------|----------------------|------------------------|----------|
| Assistant above cutoff, history tip NULL | cutoff + 5 | NULL | Deleted
|
| Assistant above cutoff, history tip below cutoff | cutoff + 20 |
cutoff - 3 | Deleted |
| Assistant below cutoff, history tip NULL | cutoff - 3 | NULL |
Preserved |
| Assistant at cutoff boundary, history tip NULL | cutoff | NULL |
Preserved |
| Assistant NULL, history tip above cutoff | NULL | cutoff + 2 | Deleted
|
| Assistant NULL, history tip at cutoff boundary | NULL | cutoff |
Preserved |
| Both step message IDs NULL | NULL | NULL | Preserved |
> Generated by Coder Agents
<details><summary>Review notes</summary>
- Run-level message IDs are below the cutoff to isolate step-level
selection.
- The assistant-above-cutoff scenario includes a second nonmatching step
to cover mixed-step deletion.
- The test uses unique model and chat names for isolation.
- `go test -v ./coderd/database -run
TestDeleteChatDebugDataAfterMessageID -count=1` passes.
</details>
Anthropic task name responses can include valid JSON followed by a
closing fence or extra text, which made `json.Unmarshal` fail with
trailing-character errors and forced fallback naming.
This updates task name JSON extraction to accept the first JSON value
after optional fences and adds regression coverage for fenced and bare
JSON with trailing content.
Anthropic rejects inline images over 5,242,880 bytes, but our upload
endpoint accepts images up to 10 MiB — so 5–10 MiB images were
reaching the provider and failing. This adds two layers of
protection: the browser resizes oversized images before upload, and
the server rejects any that still slip through before an upstream
request is issued.
Client-side resizing uses `createImageBitmap` with
`resizeWidth`/`resizeHeight` to clamp the decoded bitmap at decode
time, then iteratively shrinks on an `OffscreenCanvas` (falling back
to `HTMLCanvasElement`) until the output fits the applicable budget.
Anthropic (and Bedrock-hosted Claude — fantasy's bedrock provider is
a thin wrapper around the Anthropic client) uses a ~5 MiB budget;
other providers use a ~10 MiB budget to stay under the server cap.
Doing the resize in the browser avoids decoding attacker-controlled
image bytes in `coderd` (image-bomb DoS surface).
Server-side, `chatFileResolver` now takes a provider string and
looks up the inline-image cap via a new
`chatprovider.InlineImageByteCap`
helper; oversized `image/*` files for capped providers are rejected
with a pre-classified `chaterror` before the SDK call. The backstop
fires for older clients, direct API callers, or any image that was
committed to the composer before the user switched to a stricter
provider.
Attachments commit to composer state synchronously with a new
`"processing"` `UploadState` so paste+Enter can't dispatch before
the resize finishes; the `"uploading"` send gate now covers both
states. Dismissed-while-resizing attachments are tracked in a
`WeakSet` so a late swap can't resurrect a removed file.
Closes CODAGT-215
Closes#24090
Enhances the existing `TestFinalizeStaleChatDebugRows` test with three
missing coverage areas:
1. **Error JSON preservation**: verifies pre-existing error payloads are
not overwritten by finalization
2. **Timestamp correctness**: verifies `updated_at` and `finished_at`
match the `@now` parameter across all finalized row paths
3. **Null error preservation**: verifies finalized steps that had no
error keep a null error column
No production code changed. Test passes against Postgres.
> 🤖 Generated by Coder Agents
<details><summary>Review notes</summary>
- Enhances existing test rather than adding a new one, the existing test
was the right place
- Covers stale, orphaned, and cascade finalization timestamp assertions
- Preserves both pre-existing error JSON and null error values during
finalization
</details>
## Description
Adds automatic key failover for passthrough routes for the Anthropic and OpenAI providers. A new `keyFailoverTransport` wraps the reverse-proxy transport: centralized requests walk the configured key pool and retry with the next key on key-specific failures (401/403/429), reusing the same key-marking semantics as the bridged routes.
BYOK passthrough requests run as a single attempt with no failover.
## Changes
- New `keypool.KeyFailoverConfig` carrying the `Pool` to walk and the provider-specific closures (`IsBYOK`, `InjectAuthKey`, `MarkKey`, `BuildExhaustedResponse`).
- New `keypool.NewKeyFailoverTransport`: wraps an inner `http.RoundTripper`. Returns `inner` unchanged when `Pool` is nil, otherwise produces a transport that buffers the request body once, walks the pool per request, and replays each attempt with the next key.
- New `Provider.KeyFailoverConfig(logger)` interface method. Anthropic injects `X-Api-Key`; OpenAI injects `Authorization: Bearer ...`; Copilot returns an empty config.
- `passthrough.go` wires `NewKeyFailoverTransport` around the existing apidump middleware, so every retry attempt is recorded.
## Related Issues
Related to: https://github.com/coder/internal/issues/1446
Related to: https://linear.app/codercom/issue/AIGOV-197/aibridge-automatic-key-failover-for-bridged-and-passthrough-routes
## Follow-up PRs
- Remove dead `Provider.InjectAuthHeader` method now that all auth is applied per-attempt by `KeyFailoverTransport`.
- Bedrock multi-key support.
- Refactor provider vs interceptor config separation.
- Record the actually-used key in the interception credential hint after failover.
> [!NOTE]
> Initially generated by Claude Opus 4.7, modified and reviewed by @ssncferreira
This change uses separate http clients/transports in TestValidateToken
subtests. Previously parallel subtests of TestValidateToken shared
a http.DefaultTransport. When one subtest's httptest.Server.Close() ran in
t.Cleanup, it called http.DefaultTransport.CloseIdleConnections, which
could interrupt connection(s) used in another subtest.
## Description
Adds automatic key failover for centralized OpenAI provider, covering both chat completions and responses APIs. Same shape as the Anthropic PR: each upstream call walks the configured key pool, keys are marked **temporary** on 429 (with cooldown from `Retry-After`) and **permanent** on 401/403. Each agentic-loop iteration gets its own fresh walker so a tool-call continuation can fail over independently of the initial request.
BYOK is unchanged: BYOK requests run as a single attempt with no failover.
## Changes
- `config.OpenAI` carries a `KeyPool`. `Key` remains for BYOK Authorization Bearer set per interception.
- Chat completions blocking interceptor: walks the pool via `newChatCompletionWithKeyFailover`, marks keys on key-specific failures, returns on first success or non-failover error.
- Chat completions streaming interceptor: per-iteration walker. Pre-stream failures fail over to the next key; mid-stream errors are relayed as SSE events.
- Responses blocking interceptor: extracts `newResponseWithKeyFailover` parallel to chatcompletions.
- Responses streaming interceptor: per-iteration walker, retains the existing buffer-then-forward design.
## Related Issues
Related to: https://github.com/coder/internal/issues/1446
Related to: https://linear.app/codercom/issue/AIGOV-197/aibridge-automatic-key-failover-for-bridged-and-passthrough-routes
## Follow-up PRs
- Bedrock multi-key support.
- Refactor provider vs interceptor config separation.
- Record the actually-used key in the interception credential hint after failover.
> [!NOTE]
> Initially generated by Claude Opus 4.7, modified and reviewed by @ssncferreira
## Description
Adds automatic key failover for centralized Anthropic provider. When a key pool is configured, each upstream call walks the pool and tries keys in order until one succeeds or the pool is exhausted. Keys are marked **temporary** on 429 (with cooldown from `Retry-After`) and **permanent** on 401/403. Errors that aren't key-specific don't trigger failover. Each agentic-loop iteration gets its own fresh walker, so a tool-call continuation can fail over independently of the initial request.
BYOK is unchanged: BYOK requests run as a single attempt with no failover.
## Changes
- `config.Anthropic` carries a `KeyPool`. `Key` remains for BYOK X-Api-Key set per interception.
- Blocking interceptor: walks the pool, marks keys on key-specific failures, returns on first success or non-failover error.
- Streaming interceptor: per-iteration walker. Pre-stream failures fail over to the next key; mid-stream errors are relayed as SSE events.
- New `keypool` error types: `TransientExhaustionError` (carries soonest cooldown) and `ErrPermanentExhaustion`. Replace the prior `ErrAllKeysExhausted`.
- Error responses now consistently include the outer `"type": "error"` field.
## Related Issues
Related to: https://github.com/coder/internal/issues/1446
Related to: https://linear.app/codercom/issue/AIGOV-197/aibridge-automatic-key-failover-for-bridged-and-passthrough-routes
## Follow-up PRs
- Bedrock multi-key support.
- Refactor provider vs interceptor config separation.
- Record the actually-used key in the interception credential hint after failover.
> [!NOTE]
> Initially generated by Claude Opus 4.7, modified and reviewed by @ssncferreira
The async title-generation and turn-summary goroutines launched from
processChat run autocommit UPDATEs on the chat row after finishActiveChat
has set the chat to pending and signalWake has fired. If the row lock
from one of those UPDATEs is held while acquireLoop's processOnce runs,
AcquireChats's FOR UPDATE SKIP LOCKED skips the freshly-pending chat and
returns no rows. The wake is then consumed with no acquisition, and the
chat sits in pending until the next acquireTicker (default 1s).
Wake again after each UPDATE commits. The second wake covers the race
window without changing the transaction semantics.
Closescoder/internal#1500
## Summary
- switch the Ubuntu 26.04 dogfood image to Docker's jammy apt repository
so Docker 27 remains available
- pin `docker-ce` and `docker-ce-cli` to the Docker 27 line and keep
`containerd.io` pinned to `1.7.23-1`
- fold the containerd pin into the Docker preferences file, remove the
duplicate containerd preferences file, and hold the installed Docker
packages in the image
## Notes
Docker 28+ requires `containerd.io >= 1.7.27`, but sysbox /
Docker-in-Docker currently requires `containerd.io=1.7.23-1`, so the
image needs the older Docker 27 packages from the jammy repo.
## Testing
- Not run locally; verified the branch diff only.
Adds a persisted, draggable left sidebar width for the agents page. The
resize handle uses the same pointer-capture resize technique as the
existing right panel and clamps the expanded sidebar between 240px and
`min(520px, 50vw)`.
Updates the agents page skeleton to read the same stored sidebar width
and adds Storybook interaction coverage for resize clamping and
persistence.
Polishes the advisor tool card so the header uses a clearer lightbulb icon, inline pill metadata, wrapped/clamped question text, and a separate advice pill before the rendered response.
Adds Storybook coverage for a long advice plus 1.8k-character question peak state, including collapse/expand behavior.
Refs https://linear.app/codercom/issue/CODAGT-322/improve-advisor-icon-and-duplicate-loading-ui
<details>
<summary>Coder Agents disclosure</summary>
This PR was generated by Coder Agents.
</details>
Skips `TestExploreChatSendMessageCannotMutateMCPSnapshot` while the
chatd redesign is in flight. The test exposes a self-interrupt race in
`processChat`'s control-pubsub subscriber that is structurally fixed by
the redesign in #24444; skipping until then matches the existing
`TestSubscribeRelayEstablishedMidStream` skip in
`enterprise/coderd/x/chatd/chatd_test.go`.
Relates to https://github.com/coder/internal/issues/1493.
Zach