Commit Graph
2101 Commits
Author SHA1 Message Date
Thomas Kosiewski f923f4a525 chore!: remove the chat automations experiment (#30412)
Removes the experimental Chat Automations feature (`chat-automations` experiment) from main. This reverts the 16-PR stack that added it and adds two forward migrations that undo its schema and credential changes.

> [!WARNING]
> **Upgrade consequences**
>
> - **Automation data is deleted.** Migration 000615 drops the `chat_automations` table and deletes every queued message an automation delivered. Chats, their messages and ordinary queued messages stay. Messages lose their automation provenance (`automation_id`, `input_id`).
> - **Automation-only credentials are revoked.** Migration 000616 deletes API keys, OAuth2 codes and OAuth2 tokens whose only scopes (or allow-list entries) were `chat_automation:*`. Mixed credentials keep their other scopes. Custom roles lose their `chat_automation` permissions. OAuth2 apps that allowed only `chat_automation` scopes get the whitespace allowlist that grants nothing, as in 000595.
> - **A coordinated restart is required.** Stop all old replicas before the new version runs the migrations. Old replicas read the dropped table and columns, so a rolling deploy is not supported.
> - **Remove `--chat-max-automations-per-owner` before upgrading.** The flag, the YAML key `chat.maxAutomationsPerOwner` and the env var `CODER_CHAT_MAX_AUTOMATIONS_PER_OWNER` are removed. A deployment that still sets the flag or the YAML key will not start (`unknown flag` / `unknown option`). The env var is ignored. This includes dogfood. No release or tag contains the flag.
> - **Down migrations restore the schema only.** 000616 down and 000615 down recreate the 000611 schema with empty tables. They do not restore deleted automations, webhook secrets, provenance, queued deliveries, revoked credentials, stripped permissions or the deleted experiment rule.

## What is removed

- The `chat-automations` experiment and its stored runtime rule (`experiment_rule:chat-automations`).
- The `--chat-max-automations-per-owner` deployment option.
- The `chat_automation` RBAC resource, its member and chatd grants, and the `chat_automation:*` API key scopes.
- The automation sqlc queries, dbauthz wrappers and audit table entries.
- The management routes under `/api/experimental/organizations/{organization}/chat-automations`, and the webhook route `POST /api/experimental/chat-automations/{automation}/events` with its rate limiter.
- The chatd management service, webhook publish, schedule loop, the `manage_automations` tool, and the chatstate admission seam and queue promotion guard that only automations used.
- The Agents Automations page, editor, nav entry, chat labels, the Manage automations toggle and the API client.
- The `archived:any` chat search value that the automations "View chats" dialog added.
- `docs/ai-coder/agents/automations.md`, its manifest entry, the glossary entry and the related rows.

Reverted PRs: #30085, #30096, #30112, #30117, #30127, #30134, #30160, #30166, #30167, #30171, #30187, #30207, #30214, #30217, #30219, #30220.

## What stays, and why

- Migrations 000609, 000610 and 000611, their fixture, and the #30297 lock fix with `migration000609_test.go` stay unchanged. Deployments such as dogfood already ran them.
- New migration **000615** reverses the schema changes of 000609 and 000611 against the current schema. It uses the same short-wait lock-and-retry block as 000609, and it rebuilds `chats_expanded` from main's current definition minus the two automation columns. New migration **000616** follows the 000595 Tasks removal for credentials and recreates `api_key_scope` without the five `chat_automation:*` values. It keeps the `chat_project_memory:*` values from 000612.
- The `chat_automation` value stays in the `resource_type` enum, and `codersdk.ResourceTypeChatAutomation` stays. Old audit log rows stay readable, as with `task` after the Tasks removal.
- The 000615 delete fires the existing queue-version trigger, so `queue_version` moves to `snapshot_version` for each affected chat. `snapshot_version` is not bumped: the coordinated restart restarts every stream loop from an empty state, and the site does not read the version fields.
- `coderd/x/chatd/ARCHITECTURE.md` drops the sections the stack added and gets back its pre-stack wording, merged with the later main edits to the same file.

## Notes for reviewers

- #30379 (chat project page) merged while this PR was in progress. It passed `manageAutomationsEnabled` in `AgentCreatePage.test.tsx` and listed an `automations` route in `AgentsPageLayout.test.tsx`. Later main commits also added automation props in `ConversationTimeline.test.tsx` and `ConversationTimelineWorkingBlocks.stories.tsx`. This PR removes all of them.
- Remaining `automation` matches under `coderd`, `codersdk`, `enterprise`, `site/src` and `docs` are intended:
  - migrations 000609 to 000611, their fixture and `migration000609_test.go`;
  - the new 000615 and 000616 migrations and `migration000615_test.go`;
  - the `chat_automation` resource type in `dump.sql`, `models.go`, `codersdk/audit.go`, the generated API docs, `swagger.json` and `typesGenerated.ts`;
  - unrelated uses of the word "automation" (docs, `chatlabels_test.go`, `ratelimit.go`, `autostop.go`, `subagent_catalog.go`, `WorkspaceSharingSection.tsx`, `AgentChatPage.stories.tsx`).

Remote UAT on an isolated dev deployment passed. It upgraded a populated pre-removal database, then checked chats, queued messages, tokens, the 404 routes, the removed page and nav entry, and the experiments list.

---
_Generated with [`xum`](https://github.com/coder/xum) • Model: `anthropic:claude-opus-5-5` • Thinking: `high`_

![Agents sidebar after the upgrade, with no Automations entry](https://github.com/user-attachments/assets/bd446035-7fc8-4ca4-a735-3a4f36380738)
2026-10-11 19:57:49 +02:00
Nick Vigilante 0efb574355 feat(cli/cliui): add TextFormatFunc and OutputFormatter.WriteLine helpers (#30508)
Adds two small helpers to `cli/cliui/output.go` so commands that add
`-o/--output` don't repeat the same boilerplate, as suggested in
https://github.com/coder/coder/pull/30152#issuecomment-6043360441 by
@jeremyruppel.

- `TextFormatFunc[T]` wraps `ChangeFormatterData(TextFormat(), ...)`
with the type assertion and "expected X, got Y" error, so a command only
supplies its render function.
- `(*OutputFormatter).WriteLine` formats data and writes it to an
`io.Writer` with a trailing newline, replacing the `Format` + `Fprintln`
block.

No existing callers change in this PR. These PRs are stacked on it and
use the helpers:

- #30135 `coder tokens create`
- #30139 `coder ai-gateway keys create`
- #30145 `coder organizations members edit-roles`
- #30152 `coder provisioner keys create`

Migrating the existing callers on `main` (for example
`cli/organization.go`, `enterprise/cli/workspaceproxy.go`) is left for a
follow-up, since some of them print with `Fprint` rather than `Fprintln`
and need care to keep output identical.

Refs CLI-56

<details>
<summary>Decision log</summary>

- A global `-o` flag on the root command was ruled out because `coder
config-ssh` already uses `-o` for `--ssh-option`.
- The "one-time secret plus metadata" response structs in #30135 and
#30152 stay local to each command, since they embed different SDK types
and a shared helper would save little.
- `WriteLine` uses `Fprintln` to match the four stacked PRs and most
existing callers.
It was renamed from `Write` during review so the trailing newline is
visible at call sites and the method is not confused with `io.Writer`.

</details>

> [!NOTE]
> This PR was generated by Coder Agents on behalf of @nickvigilante.
2026-10-08 19:41:31 +00:00
Nick Vigilante a8303a8f2b feat!: add output formats to coder schedule start, stop, and extend (#30136)
`coder schedule start`, `stop`, and `extend` printed the updated
schedule as a fixed table with no `-o` or `-c`, while `coder schedule
show` has both.

All four commands now share one formatter with the same row type,
default columns, `-o table|json`, and `-c`, so the output of a change
and of `show` is interchangeable.

**Behavior change:** Adds `-o` and `-c` to the three commands.
The default table output changes slightly: the WORKSPACE column no
longer includes the favorite-star prefix, matching `schedule show`.
Other columns are unchanged.

<details>
<summary>Decisions</summary>

- Previously these commands rendered `WorkspaceListRow`, which prefixes
the workspace name with a favorite icon. Using `scheduleListRow` for
consistency with `schedule show` drops that prefix.

</details>

Closes [CLI-15](https://linear.app/codercom/issue/CLI-15)

_This PR was generated by Coder Agents on behalf of @nickvigilante._
2026-10-08 13:07:17 -04:00
Nick Vigilante 98bcc3afb7 fix!: remove the unused --yes flag from coder autoupdate (#30149)
`coder autoupdate` has registered `-y/--yes` since #10438 but never
prompts, so the flag did nothing.
This PR removes it.

**Breaking change:** `coder autoupdate` no longer accepts `-y/--yes`.
Scripts that pass it now fail with `unknown shorthand flag: 'y'`.
To fix them, drop the flag; it never had an effect.

<details>
<summary>Decisions</summary>

- The first version kept `-y` as a hidden no-op so existing scripts
wouldn't break.
- Reviewers preferred removing it outright as a breaking change, since a
deprecation period would end in the same removal and the hidden flag is
cruft.
- The title uses `fix!:` and the PR carries `release/breaking`, so the
release tooling lists it under breaking changes.
- `cliui.HiddenSkipPromptOption()` landed on `main` through the
organizations create PR and stays there for that command; this PR no
longer touches it.
- The test now pins the new behavior: passing `-y` returns an unknown
flag error.

</details>

Closes [CLI-33](https://linear.app/codercom/issue/CLI-33)

_This PR was generated by Coder Agents on behalf of @nickvigilante._
2026-10-08 11:55:29 -04:00
Bobby Ho a115896a58 feat: add script resource address and dependencies to Script (#30451)
feat(provisionersdk): add script dependency proto fields

Add optional `resource_address` and `dependencies` fields to `Script`
for transporting resolved script ordering. Dependencies identify a
prerequisite by Terraform address and require `SUCCESS` (successful
exit) or `COMPLETION` (finish regardless of outcome).

- Bump provisionerd API from v1.19 to v1.20, regenerate Go/TypeScript
  bindings, and update the CLI golden output.
- Reuse the existing resource tree for workspace-agent and devcontainer
  scripts; no changes to `GraphComplete` or the provisionerd runner.
- Add round-trip tests for both script types and verify unset fields
  preserve existing wire encoding.

The fields remain unused in this PR. Follow-up provisioner changes
populate them; PLAT-548 reads and persists them in coderd.

**Release requirement:** Ship with the PLAT-545 integration stack,
targeting 2.39, so no daemon advertises v1.20 without sending ordering.
Hold or revert this PR before the release cut if that stack misses it.

Refs PLAT-547.
2026-10-07 22:48:21 +00:00
dylanhuff-at-coderandNick Vigilante b4d51012da feat: allow template version selection for workspace builds (#30199)
Add named template version selection to `coder start` and `coder stop`,
and independent stop/start selection to `coder update` with
`--stop-template-version` and `--template-version`. Stop still defaults
to the workspace's current version, and update starts with the active
version by default. Explicit selections use the selected version for
parameter preparation and builds, keep server permissions authoritative,
and never silently fall back to the active version.

Classic stop builds retain previous parameter values omitted by the
selected stop version, so a later start can reuse them across separate
commands or an interrupted update. Declared parameters still follow
normal validation, and start builds still discard removed parameters.
Update now returns an error and retry guidance when a start apply fails
after stopping, including unflagged updates that previously returned
success.

Closes #29100


[PLAT-616](https://linear.app/codercom/issue/PLAT-616/cli-allow-specifying-a-template-version-for-stopstart-builds-and-both)

---------

Co-authored-by: Nick Vigilante <nickvigilante@users.noreply.github.com>
2026-10-07 13:17:12 -07:00
Jakub Domeracki b4fb7901e6 docs: clarify that disable-password-auth also applies to owners (#30362)
The `--disable-password-auth` help text says owners can still sign in
with a password "to avoid potential lock out", and a comment in
`coderd/userauth.go` says the same. Neither is true. #6193 removed the
owner bypass on purpose, so the flag blocks password login for everyone.
The recovery advice was also wrong. It named a command that doesn't
exist (`create-admin`), and any admin that `create-admin-user` creates
while the flag is on still can't sign in.

This PR changes docs and comments only:
- Rewrites the flag description to say it applies to owners too and how
to recover access. Regenerated the CLI and config reference docs and the
golden files.
- Fixes the misleading comment in `coderd/userauth.go`.
- Adds the same warning and recovery steps to
`docs/admin/users/password-auth.md`.

> [!NOTE]
> Generated by Coder Agents on behalf of @jdomeracki-coder.
2026-10-07 19:03:46 +02:00
Michael Suchacz 7c39819665 fix(cli): close coderd on server startup errors (#30280)
`coder server` closes the coderd API only in its graceful shutdown
sequence. Every startup error returned after `newAPI` (creating
aibridged, registering Prometheus metrics, writing the config URL,
creating provisioner daemons, notifying systemd) skipped that close and
leaked the whole coderd instance: tailnet and wireguard-go, gvisor,
metricscache, dbrollup, cryptokeys, chatd, and pubsub queues.
`TestServer/Logging` hits this when test cleanup cancels startup while
aibridged is being created. The test helper ignores the resulting
`context canceled` error, so only the package-level goleak check in
`cli` fails.

The fix closes the API in a deferred `sync.OnceValue` registered right
after `newAPI`, and the shutdown sequence calls the same function. The
guard is needed because the closer is not safe to call twice: the AGPL
API returns "API already closed", and the enterprise closer closes each
component again. The normal shutdown path is unchanged.

## Testing

The new `TestServer/StartupErrorNoLeak` fails startup after coderd is
created by putting a directory at the config URL path.

- Red (test only): `go test ./cli -run
'^TestServer$/^StartupErrorNoLeak$' -count=1` exited 1 in 3 of 3 runs
with `goleak: Errors on successful test run: found unexpected
goroutines` (coderd frames) while the subtest passed.
- Green (with the fix): the same command passed in 3 of 3 runs.
- `TestServer/Logging` alone, 20 runs from one test binary: 20 of 20
passed with no goleak failure, including 3 runs that hit the `create
aibridged` startup error.
- `go test ./cli -count=1` passed in 4 of 4 runs, and `go vet ./cli` is
clean.

Found in the CI of #30272 (`test-go-pg (ubuntu-latest)`).

> Xum, an AI coding agent, implemented, tested, and opened this PR on
behalf of @ibetitsmike.
2026-10-07 13:39:31 +02:00
McKayla はな dad9f733f4 chore: improve code vpn-daemon run help text (#30331) 2026-10-06 11:52:40 -06:00
Nick Vigilante 237616a87d fix!: make --yes skip the rename confirmation prompt (#30151)
`coder rename` registers `-y/--yes`, but its confirmation asks you to
type the workspace name, and `cliui.Prompt` only honors `-y` for yes/no
prompts.
So `coder rename ws new -y` still blocked on input.

With `-y`, the type-the-name prompt is now skipped.
The data-loss warning is still printed.

**Behavior change:** With `-y`, the rename happens without reading
stdin.
Without `-y`, the prompt is unchanged.

Closes [CLI-35](https://linear.app/codercom/issue/CLI-35)

_This PR was generated by Coder Agents on behalf of @nickvigilante._
2026-10-06 10:08:14 -04:00
Callum Styan 15616728d5 refactor: move coderd/x/nats to coderd/nats 2026-10-05 16:52:36 -07:00
Paweł Banaszewski ffc7dadfe2 refactor: use typed AI Gateway actor identity (#30258)
Adds typed AI Gateway actor identity, including the API key ID used by
recording.

Recorders now read the authenticated API key ID from each call context
instead of capturing request identity at construction time. This allows
one recorder to be reused safely across sequential and concurrent
requests. Includes context helper coverage and recorder regression
tests.

Generated by Coder Agents on behalf of @pawbana.
2026-10-05 14:00:15 +00:00
Asher cc0c868c55 feat: add client_session_id to ssh connections (#29701)
Add a new /tcp endpoint on the agent that is used to pass along
the client session ID (with the existing baggage mechanisms). The agent
then can upgrade that into the desired connection type, based on the
port. For now that is only the standard SSH port and reconnecting pty.

The `ssh`, `ssh --stdio`, `ssh --stdio` over Coder Connect, `exp rpty`,
and web terminal paths are all updated to use this new endpoint, and
fall back to the old method when the endpoint is not available.

That ID is added to the log context, and is also used to populate the
session ID column in the connection logs table.

There are some logs related to a connection that will not be tagged with
the ID (like the "accepted connection" line) because they happen before
the http handler gets to run.
2026-10-02 16:40:24 -08:00
Thomas Kosiewski 9316d67440 feat: add chat automations management API behind chat-automations (#30112)
Adds the management API for chat automations under
`/api/experimental/organizations/{organization}/chat-automations`,
behind the new user-scoped `chat-automations` experiment. This is the
first of two PRs for RFC step 3; nothing publishes automation messages
yet.

- **Experiment:** `chat-automations` is in `ExperimentsKnown`,
`ExperimentsUserScoped`, and the ruleguard list in `scripts/rules.go`.
Every route asks `chatd.AutomationsEnabled(ctx, api.ExperimentEvaluator,
callerID)`, which fails closed on a missing subject or evaluator; the
routes stay mounted and return 404 when the experiment is off for the
caller, so a runtime rule for one user applies on the next request
without a restart. There is no `/api/v2` alias.
- **Routes:** list, create (201), get, `PATCH`, delete (204), with
Swagger annotations and audit entries. List returns the automations in
the organization the caller can read (members see their own,
organization admins see all).
- **Service** (`coderd/x/chatd/automations.go`): create validates kind
and target shape, a non-empty prompt, a five-field cron and IANA time
zone, model access, and that an `existing_chat` target is a root chat of
the owner in the same organization. It defaults `when_busy` to `skip`
for schedules and `queue` for webhooks and computes the first schedule
cursor. `PATCH` runs the same per-field validation on each changed
field, rejects `kind` and `target_mode`, bumps `schedule_revision` on
prompt, cron, or time zone changes, and recomputes the cursor on cron or
time zone changes.
- **Owner-only:** only the owner can `PATCH` an automation, even with
organization-admin or site-owner roles (403); admins can list, read, and
delete.
- **Secrets:** a webhook's secret (`coder_automation_` prefix) is
returned once in the create response. Only its SHA-256 is stored, and it
appears in no list, get, audit, or log output.
- **Cap:** new option `--chat-max-automations-per-owner` (default 50)
caps automations per owner across all organizations; creating one more
returns 409. Creates serialize on a per-owner advisory lock.
- **Cron:** `cron.Standard(spec, timeZone)` accepts any standard
five-field expression (the existing `Weekly` rejects day-of-month and
month restrictions). Responses include the next five run times of an
enabled schedule.

PR 3b (stacked on this one) adds disable and re-enable with the queue
cutoff, makes delete run disable first, and adds secret rotation and the
schedule preview route. Until then, the queue promotion guard from
#30096 discards queued rows whose automation was deleted.

Dogfood UAT ran on a real dev instance and passed:
create/list/get/patch/delete via curl, owner-only 403s for an
organization admin and the site owner, the cap (including concurrent
creates and cross-organization counting), a runtime rule turning the
experiment off for one user without a restart, and `/api/v2` 404s.

Part of CODAGT-1209 (PR 3b closes it)
Part of CODAGT-1135
Refs CODAGT-1217

Stack: PR 3 of the Automations for Coder Agents stack, on top of #30096
(which is on top of #30085).

> This PR was written by an AI agent (xum) and reviewed by the agent's
own test audit and UAT. A human has not reviewed it yet.

---
_Generated with [`xum`](https://github.com/coder/xum) • Model:
`anthropic:claude-opus-5-5` • Thinking: `high`_

---------

Signed-off-by: Thomas Kosiewski <tk@coder.com>
2026-10-02 15:54:49 +00:00
Nick Vigilante 1ad46e48ac feat: report port-forward sessions as app usage (#30156)
`coder port-forward` reported usage with the deprecated
`UpdateWorkspaceUsageContext`.
It now reports a `port_forwarding` session, like `coder ssh` and the VS
Code SSH path, but only while a forwarded TCP connection is open.

**Behavior change:** CLI output is unchanged.
Port-forward sessions previously only bumped the workspace's last-used
time.
With the hidden `workspace-usage` experiment enabled, the server now
also records `port_forwarding` session counts under a new
`port_forwarding` family, which the deployment banner totals under
**Other**.
Without the experiment, `postWorkspaceUsage` returns 204 before reading
the body, so nothing changes.

<details>
<summary>Decisions</summary>

- Added `codersdk.UsageAppNamePortForwarding` (`port_forwarding`) rather
than reusing `ssh`, so port-forwarding isn't counted as interactive SSH.
The name matches the arbitrary app names RFC's connection-method name.
- Registered a `port_forwarding` family in `codersdk/appname.go`.
Families resolve at read time, so no data migration is needed. Template
insights list apps individually; telemetry, metricscache, and agentstats
gauges are legacy fixed fields and stay as they are. The Prometheus
insights family is left to #29906.
- Only TCP connections count. UDP forwards never see EOF, so one
datagram would keep the session active until exit.
- Reports once at startup, like the old call, then every minute and once
on shutdown. An idle tick sends an empty body, which still bumps
last-used.

</details>

Closes [CLI-31](https://linear.app/codercom/issue/CLI-31)

_This PR was generated by Coder Agents on behalf of @nickvigilante._


<details>
<summary>CI note: unrelated test-go-race-pg flake</summary>

The `test-go-race-pg` job failed on a data race between
`tailnet.(*configMaps).updatePeerLocked` and slog formatting of the same
peer value, which does not touch port-forward code. The job was rerun.

</details>
2026-10-02 11:05:48 -04:00
Paweł Banaszewski 7e64b46721 fix: drain proxy requests before closing DRPC (#29724)
Keep DRPC available while admitted AI Gateway proxy requests drain,
rather than canceling the DRPC connection lifecycle at the start of
shutdown.
After shutdown reject new handler acquisition immediately.
Once requests finish or the shutdown context expires, cancel remaining
proxy requests and the DRPC lifecycle. Shutdown stops waiting for the
DRPC connection loop when its deadline expires.
Reconnects skip initialization once a backend is selected.

Generated by Coder Agents on behalf of @pawbana.
2026-10-01 12:25:11 +00:00
Nick Vigilante 0d1225694e feat: show coder sharing in CLI help and reference docs (#30141)
`coder sharing` has been hidden since it was added behind an experiment,
although Shared Workspaces is GA and the prose docs already use it.

This removes `Hidden: true`, regenerates `docs/reference/cli/sharing/*`,
and adds the pages to the docs manifest.
The shared workspaces guide and the persistent shared workspaces
tutorial now link to the new reference.

**Behavior change:** `coder sharing` and its subcommands now appear in
`coder --help`.
The commands themselves are unchanged.
The `sharing remove --user/--group` help text previously said "to share
the workspace with"; it now says "to remove from the workspace".

<details>
<summary>Decisions</summary>

- Follow-up not in this PR: `DELETE /workspaces/{workspace}/acl` has no
CLI caller.

</details>

Closes [CLI-22](https://linear.app/codercom/issue/CLI-22)

_This PR was generated by Coder Agents on behalf of @nickvigilante._
2026-09-30 21:29:51 +00:00
Nick Vigilante 46265fc361 fix: hide the unused --yes flag on coder organizations create (#30144)
`coder organizations create` registered `-y/--yes` but has had no
confirmation prompt since the org-delete prompt was removed in #16972.

`-y` is now a hidden no-op: it's gone from help and the reference docs
but still accepted, so existing scripts don't break.

**Behavior change:** `-y/--yes` no longer appears in `coder
organizations create --help`.
Passing it still works.

<details>
<summary>Decisions</summary>

- Adds `cliui.HiddenSkipPromptOption()`. The autoupdate PR adds the same
helper, so whichever merges second will need a trivial rebase.

</details>

Closes [CLI-26](https://linear.app/codercom/issue/CLI-26)

_This PR was generated by Coder Agents on behalf of @nickvigilante._
2026-09-30 14:57:29 -04:00
Nick Vigilante b6aab35a7b feat!: confirm before coder update restarts a running workspace (#30150)
`coder update` stopped and restarted a running workspace without asking
and had no `-y`, while `coder restart` and the dashboard's update dialog
both confirm first.

It now asks for confirmation before stopping a running workspace, using
the dashboard's wording that updating can delete non-persistent data.
`-y` skips it, and also answers the existing "Enqueue another stop?"
prompt.

**Behavior change:** **New prompt; can break scripts.** `coder update`
on a running workspace without `-y` now blocks on stdin, or fails if
stdin is closed.
Stopped workspaces don't prompt.
Parameter prompts such as `--always-prompt` aren't affected by `-y`.

The [Update
workspaces](https://coder.com/docs/user-guides/workspace-management#update-with-the-cli)
guide now notes the prompt and `-y`.

<details>
<summary>Decisions</summary>

- Mirrors `coder restart`'s prompt pattern.
- Doc-check suggested a row in the 2.29 to 2.34 ESR upgrade guide. This
change ships after 2.34, so it's documented in the user guide instead.

</details>

Closes [CLI-34](https://linear.app/codercom/issue/CLI-34)

_This PR was generated by Coder Agents on behalf of @nickvigilante._
2026-09-30 14:31:43 -04:00
Nick Vigilante 04b2af25ca fix(cli): offer a None option in the coder create preset prompt (#30068)
When a template has presets but no default preset, `coder create`
prompts for a preset, but the prompt listed only the template's presets.
The only way to skip a preset was `--preset none`, while the dashboard
lists **None** first.

This adds a **None** entry at the top of the prompt.
Selecting it behaves like `--preset none`: the CLI sends no preset ID
and prints `No preset applied.`
If a preset's label is also `None`, the no-preset entry is labeled `None
(no preset)` so both stay selectable.
Default preset auto-selection is unchanged.

> [!NOTE]
> "No preset ID sent" doesn't always mean the build records no preset.
> When a create request has no preset ID, coderd matches the parameter
values against the template's presets (`prebuilds.FindMatchingPresetID`)
and records any exact match, as `--preset none` and the dashboard's None
option already do.

Fixes ENG-3407

<details>
<summary>Validation and decision log</summary>

### Validation

Tested at e5f52ffe4b against Postgres.

- `TestCreate`, `TestCreateWithPreset`, and the new
`TestPresetSelectOptions` pass against Postgres 13.
- Ran a dev server from this branch (`./scripts/develop.sh
--port-offset`) and drove `coder create` through a real PTY against a
template with two presets and no default:
  - The menu lists `None`, `eu-preset: Deploys to EU`, `us-preset`.
  - `None` prints `No preset applied.`, then prompts for parameters.
  - Selecting either preset applies its parameters.
- `--preset none` and the prompt's `None` produce identical
`workspace_builds.template_version_preset_id` values.
With `region=us`, both record the matching `us-preset` because of
server-side preset matching.
- Template with a preset literally named `None`:
- The dashboard shows `None` twice. Each entry has its own internal
value, so both work, but they look identical.
- The first commit on this branch showed `None` twice in the CLI too,
and both entries meant "no preset". That made the `None` preset
unreachable from the prompt, which is a regression from `main`.
- The second commit fixes that: the menu shows `None (no preset)`,
`None`, `us-preset`, and selecting `None` applies the preset
(`region=eu`).

### Decisions

- **Label `None` to match the dashboard.** Switch to `None (no preset)`
only when a preset's label collides, so the common case stays identical
to the dashboard.
- **Out of scope: `--preset None` flag.** The flag matches `none`
case-insensitively before any preset name, so a preset named `None` or
`none` can't be selected by flag. That's unchanged from `main`.
- **Out of scope: default presets.** The dashboard preselects the
default preset but still lets you pick another preset or None. The CLI
applies the default without prompting. That behavior needs a separate
decision.

</details>

---

_Generated by Coder Agents on behalf of @nickvigilante._
2026-09-30 14:18:26 -04:00
Nick Vigilante c9813d97f7 fix: swap description and command in coder organizations show examples (#30154)
The `coder organizations show` help examples had their `Description` and
`Command` fields swapped, so `--help` and
`docs/reference/cli/organizations/show.md` read backwards.

This swaps them back, tidies the descriptions, and regenerates the help
snapshot and reference page.

**Behavior change:** Help text and generated docs only.

Closes [CLI-38](https://linear.app/codercom/issue/CLI-38)

_This PR was generated by Coder Agents on behalf of @nickvigilante._
2026-09-30 13:00:38 -04:00
Nick Vigilante 41e698edc3 fix(cli)!: match the exact workspace in coder schedule show (#30153)
`coder schedule show owner/ws` dropped the owner and searched
`name:<ws>`, and the workspace search matches names as substrings.
As an owner, `schedule show alex/ui` returned every workspace containing
`ui`; even `schedule show ui` also matched `ui-2`.

When a workspace argument is given, the command now resolves that exact
workspace with `ResolveWorkspace`, like other `[owner/]workspace`
commands.
`--search` and `--all` are unchanged.

**Behavior change:** `schedule show <workspace>` returns only that
workspace.
A workspace that doesn't exist, or that you can't access, now exits with
an error instead of printing `No schedules found.` and exiting 0.

<details>
<summary>Decisions</summary>

- The fix is CLI-only; the server's substring name search is left as is.

</details>

Closes [CLI-37](https://linear.app/codercom/issue/CLI-37)

_This PR was generated by Coder Agents on behalf of @nickvigilante._
2026-09-30 12:40:27 -04:00
Nick Vigilante e6774c2602 feat: add --column to coder users show (#30143)
`coder users show` offered `-o table|json` but was the only table
command without `-c/--column`, because it uses a custom key/value
format.

The custom format now registers `-c` and filters its rows.
Column names match `users list` where they overlap: `id`, `username`,
`name`, `email`, `status`, `created at`, `roles`, `organizations`.

**Behavior change:** Additive.
The default output is byte-for-byte unchanged.

<details>
<summary>Decisions</summary>

- Switching to `cliui.TableFormat` would have turned the vertical
key/value layout into a horizontal table, so the existing renderer is
kept and filtered instead.

</details>

Closes [CLI-25](https://linear.app/codercom/issue/CLI-25)

_This PR was generated by Coder Agents on behalf of @nickvigilante._
2026-09-30 12:38:45 -04:00
Nick Vigilante dde500a646 feat: add --yes to coder users suspend and activate (#30155)
`coder users suspend` and `coder users activate` always prompt but,
unlike other commands with a confirmation prompt, had no `-y/--yes`.

Both now accept `-y`.
The suspend help also says that suspending stops the user's running
workspaces and that activating doesn't restart them.

**Behavior change:** Additive.
New `-y` flag; without it, both commands prompt as before.

<details>
<summary>Decisions</summary>

- The workspace note is based on the lifecycle executor, which
auto-stops a suspended user's running workspaces; activation doesn't
restart them.

</details>

Closes [CLI-23](https://linear.app/codercom/issue/CLI-23)

_This PR was generated by Coder Agents on behalf of @nickvigilante._
2026-09-30 12:38:14 -04:00
Nick Vigilante cf60604a75 feat!: confirm role changes in coder users edit-roles (#30140)
`coder users edit-roles --roles ...` replaces the user's whole set of
site roles, so granting one role could silently remove others, including
owner.
It registered `-y` but never prompted, and `-y` didn't skip the
interactive role picker.

It now prints the roles being added and removed and asks for
confirmation; `-y` skips it.
The `--roles` help says it replaces the full set.

**Behavior change:** **New prompt; can break scripts.** `edit-roles
--roles ...` without `-y` now prompts.
`-y` without `--roles` now fails with an error instead of opening the
picker.
If the requested roles match the current roles, it prints `No role
changes for <user>.` and doesn't prompt.

<details>
<summary>Decisions</summary>

- No-op requests skip the prompt so idempotent scripted calls stay
quiet.

</details>

Closes [CLI-21](https://linear.app/codercom/issue/CLI-21)

_This PR was generated by Coder Agents on behalf of @nickvigilante._
2026-09-30 12:36:30 -04:00
Nick Vigilante 302edffd87 feat: add JSON output to coder server create-admin-user (#30138)
`coder server create-admin-user` is mostly used in automation but only
printed text.

It now accepts `-o text|json`.
JSON contains the created user's ID, username, email, roles, and
organization memberships.

**Behavior change:** Additive.
The default text output is unchanged and still goes to stderr.
JSON goes to stdout so it can be piped.
The password is never included in any format.

<details>
<summary>Decisions</summary>

- The command talks to the database directly, so the JSON uses a small
local response type instead of a codersdk type.
- JSON goes to stdout while text stays on stderr, matching how other
commands emit machine-readable output.

</details>

Closes [CLI-17](https://linear.app/codercom/issue/CLI-17)

_This PR was generated by Coder Agents on behalf of @nickvigilante._
2026-09-30 12:36:05 -04:00
Andrew Aquino 4a67ac3306 feat: buffer CLI session logs and flush on connection failure (#29752)
## What

Adds in-memory log buffering for CLI diagnostics so the debug detail
leading up to a failure is available without logging debug during normal
operation. Below-level entries are kept in a slog flight recorder
(`v3.2.0`) and emitted only when a command returns an error. This
implements CLI-4 (connection-log RFC req 13). Depends on the
`cdr.dev/slog/v3` `v3.2.0` bump (already on `main`).

## How

- **Root command** (`cli/root.go`):
- New global `--log-buffer-size` / `CODER_LOG_BUFFER_SIZE` (default
`10`, `0` disables, clamped to `10000`), shared by all commands instead
of duplicated per command.
- `bufferedLoggerMiddleware` runs the invocation logger at `Info` (or
`Debug` with `-v`) with a flight recorder, and on a returned error calls
`Flush` to replay the buffered history to stderr. `Flush` does not add a
duplicate error line; the returned error is still rendered by the normal
formatter.
- Opt-in via an `annotationBufferedLogger` annotation so it does not
clobber the ~17 commands that build their own logger on `inv.Logger`
(which would otherwise double-log). Future commands opt in with one
line.
- `start` / `stop` / `update`: opt into the middleware and just
`client.SetLogger(inv.Logger)` so codersdk request/response debug is
captured and flushed to stderr on failure. They **no longer write
session log files** and dropped their per-command `--log-dir` /
`--log-buffer-size`.
- `ssh`: keeps its own file logger (writing to stderr would corrupt its
stdio stream), now using the global buffer size and the shared
`bufferedLogger` helper. It retains `--log-dir` / `CODER_SSH_LOG_DIR`,
stdio/PID filename handling, and file pruning (50 most recent).
- `restart` is excluded, matching the `client_session_id` opt-in set
(DEVEX-664).

## Default behavior change

`start` / `stop` / `update` now emit buffered diagnostics (recent
below-level entries, e.g. sdk request/response) to **stderr** when they
fail; on success they stay quiet. `coder ssh` continues to write a
per-connection diagnostic file by default under the XDG state dir
(`~/.local/state/coder/logs`), pruned to the 50 most recent. Opt out of
buffering with `--log-buffer-size 0`.

## Review changes (Asher)

- Shared logger construction: the level is computed once and the flight
recorder is always attached, in a single `bufferedLogger` helper used by
ssh and the workspace commands.
- Workspace commands print to normal output instead of a log file, and
the default buffer is `10` to avoid flooding the terminal on failure.
- The buffer lives on the root command / logger (global option +
middleware) rather than being wired into each command.

## Alignment with #29671 (agent)

Same core model: `Info` default with a flight recorder, `-v` for full
debug, and the shared `--log-buffer-size` semantics (`0` disables).
Differences: the agent writes to a file and defaults the buffer to
`1000`, while the CLI prints to stderr and defaults to `10`; both now
rely on slog's auto-flush-on-error (the CLI additionally flushes via the
middleware on a returned error).

> [!NOTE]
> The agent PR (#29671) adds an agent-scoped `--log-buffer-size` /
`CODER_AGENT_LOG_BUFFER_SIZE` on the hidden `coder agent` command. Once
both land, the global `--log-buffer-size` here and the agent's own flag
will need reconciling on `coder agent`. Flagging for whichever merges
second.

## Testing

`cli.TestClampLogBufferSize`, `TestPruneSessionLogs`,
`TestDefaultSessionLogDir`, and `TestBufferedLogger_*` cover clamping,
ssh log retention, the default state dir, buffer-until-error, the
middleware's flush path, and the verbose escape hatch.
`start`/`stop`/`update`/`ssh` command tests pass. Golden `--help` output
and `docs/reference/cli` regenerated. (`TestSSH/Env` is a pre-existing
interactive-PTY test that does not run in all sandboxes; it fails
identically on `main` here.)

<details>
<summary>Rollout context</summary>

Cross-repo sequence: coder/slog flight recorder (released as v3.2.0) ->
go.mod bump in coder/coder (#29600, merged) -> CLIENT-672 (agent,
#29671) and this PR (CLI-4), which are independent consumers and can
merge in any order.
</details>

---
<sub>Opened by Coder Agents on behalf of @aqandrew.</sub>
2026-09-29 22:27:30 +00:00
Andrew Aquino 6487425885 feat: buffer agent debug logs and flush on connection failure (#29671)
## What

Runs the workspace agent logger at `Info` with a slog flight recorder
(`v3.2.0`) instead of unconditionally at `Debug`. Entries below the
current log level are kept in a rolling in-memory ring and flushed to
the sinks when the agent loses its connection to coderd (and before any
`Error`+ log), so the detail leading up to a failure is available
without writing debug logs during normal operation.

Implements CLIENT-672 (connection-log RFC req 13) for the workspace
agent. Depends on the `cdr.dev/slog/v3` `v3.2.0` bump (already on
`main`).

## How

- `cli/agent.go`:
- The main agent logger is
`AppendSinks(...).Leveled(level).FlightRecorder(flightRecorderSize)`,
where `level` is `Info` by default and `Debug` under the global `-v` /
`CODER_VERBOSE`. The recorder is always attached; at `Debug` it is
inert, which keeps the setup correct if the configured level changes in
the future. The reaper logger is unchanged.
- New `--agent-flight-recorder-size` /
`CODER_AGENT_FLIGHT_RECORDER_SIZE` (default `1000`, `0` disables the
flight recorder), so the ring size is configurable per RFC req 13. The
agent flag is distinct from CLI-4's global `--flight-recorder-size`
(#29752) so the two don't collide when `coder agent` inherits the root
global.
- `workspaceAgent` is a `RootCmd` method so it reuses the root command's
global verbose flag instead of defining an agent-specific one.
- `agent/agent.go` (`runLoop`): on a lost coderd connection (the
`io.EOF` disconnect and the generic run error) the agent calls
`logger.Flush(ctx)` to emit the recorded history and then logs at
`Warn`. These are `Warn` rather than `Error` because reconnectable
disconnects are expected and `slogtest` fails tests on any `Error` log;
the manual flush is what surfaces the recorded detail. Graceful shutdown
paths (`ctx` canceled, agent closed) are not flushed.
- Tests: `agent.TestAgent_FlushFlightRecorderOnDisconnect` covers the
flush-on-disconnect path; `cli.TestWorkspaceAgent/DisabledServers` runs
with verbose since its debug messages are otherwise recorded. Golden
`--help` output regenerated.

## Default log verbosity change

This changes the default agent-log verbosity. Today `coder-agent.log`
receives `Debug` unconditionally; with this change, normal operation
logs `Info`, and recorded below-level entries are emitted only when the
agent loses its coderd connection (via a manual flush) or before any
`Error`+ log, or always under `-v` / `CODER_VERBOSE`. This is the
intended behavior of the feature: a quiet steady state with full detail
retained for diagnosing failures. Operators who want the previous
always-debug output can run with `-v` / `CODER_VERBOSE`.

## Alignment with #29752 (CLI)

This mirrors the CLI-4 (CLI log buffer) design so the two
implementations agree:

- **Level model:** Info by default, debug available on demand via the
global `-v` / `CODER_VERBOSE`.
- **Flight recorder size:** default `1000`, `0` disables. CLI-4 exposes
a global `--flight-recorder-size`; the agent uses a distinct
`--agent-flight-recorder-size` / `CODER_AGENT_FLIGHT_RECORDER_SIZE` (the
`CODER_AGENT_*` env convention) so the inherited global and the agent
flag coexist without collision.
- **Flush trigger:** the recorded history is emitted before any `Error`+
log via slog's auto-flush. The agent additionally flushes manually on a
lost coderd connection, which it logs at `Warn` (reconnectable, and
`Error` would fail `slogtest`-based tests); the CLI's failures are
fatal, so it relies on the auto-flush-on-error.

<details>
<summary>Rollout context</summary>

Cross-repo sequence: coder/slog flight recorder (released as v3.2.0) ->
go.mod bump in coder/coder (#29600, merged) -> this PR (CLIENT-672,
agent) and CLI-4 (CLI), which are independent consumers and can merge in
any order.
</details>

---
<sub>Opened by Coder Agents on behalf of @aqandrew.</sub>
2026-09-29 14:24:00 -07:00
Cian Johnston 9fa17a12ef fix: strip client actor headers unconditionally (#30037)
Strip all client-supplied headers matching `IsActorHeader()` in
`PrepareClientHeaders`, regardless of `send_actor_headers`. This
reserves the case-insensitive `X-AI-Bridge-Actor` prefix for identity
injected by the gateway. Remove the redundant deletion of the three
standard actor headers from `BuildUpstreamHeaders`.

Generated with Coder Agents on behalf of @johnstcn.
2026-09-29 20:26:17 +01:00
Cian Johnston 8742d2484c feat: add opt-in AI Gateway actor email headers (#29930)
Add opt-in forwarding of the authenticated Coder account email through
`--ai-gateway-actor-header-meta-email`,
`CODER_AI_GATEWAY_ACTOR_HEADER_META_EMAIL`, or
`ai_gateway.actor_header_meta_email`. The option defaults to empty:
enabling `send_actor_headers` alone does not forward email. Set a
destination header explicitly to opt in; accounts without an email omit
it.

Carry authenticated email through the Gateway authorization response and
strip client-supplied standard/configured email headers before applying
trusted identity. The additive protocol field joins unreleased Gateway
API v1.3 alongside workspace attribution. Includes email configuration
precedence, authorization compatibility, captured upstream request
tests, and matching documentation.

Refs https://linear.app/codercom/issue/AIGOV-601

Generated with Coder Agents on behalf of @johnstcn.
2026-09-29 20:07:49 +01:00
Cian Johnston 05d1fa8a4c feat: configure AI Gateway actor identity headers (#29929)
Allows configuring AI Gateway actor ID and username headers independently through
`--ai-gateway-actor-header-id` and
`--ai-gateway-actor-header-username` or associated environment
variables and YAML settings. 

- Defaults to existing header name, an empty string disables that attribute. 
- Existing `send_actor_headers` gate remains off by default.
- Validates header names, case-insensitive collisions, and protected
destinations for embedded and standalone gateways.
- Strips standard and configured identity destinations before injecting authenticated values.
- Adds coverage for configuration precedence, captured upstream request, and generated help/API documentation.

Refs https://linear.app/codercom/issue/AIGOV-601

Generated with Coder Agents on behalf of @johnstcn.
2026-09-29 18:22:59 +01:00
Thomas Kosiewski 12be685b59 feat: add experimental API and CLI to manage experiment rules (#29971)
Slice 4 of 4 for runtime experiment rules: an experimental management
API and a hidden CLI, so admins can target, widen, or kill a user-scoped
experiment at runtime without a redeploy. This PR also adds a
startup-default note on the Deployment Overview page and a docs section.

## What's in it

- **API** under `/api/experimental/experiments/rules` (behind
`apiKeyMiddleware` and `httpmw.NoStore`):
- `GET` returns, per user-scoped experiment, the startup default and the
stored rule (mode, condition, revision, actor, time). Stored rules for
other experiments follow, marked `ignored`.
  - `PUT /{experiment}` takes `{mode, condition?, expected_revision}`.
- **Write path:**
- It authorizes read and update on `DeploymentConfig` before parsing
anything. A denied request gets 403 and is not audited.
- It validates next. An invalid condition returns 400 with the full CEL
diagnostics, is not audited, and leaves the rule unchanged.
- It then starts the audit and calls `experiments.WriteRule`. A stale
revision returns 409 and a transaction error returns 500; both are
audited with an empty diff.
- An identical write returns 200, creates no audit entry, and keeps the
revision.
- The audit export happens after commit. If it fails, that is logged and
the committed change stands (the existing audit contract).
- **SDK:** `ExperimentalClient.ExperimentRules` and
`ExperimentalClient.PutExperimentRule`.
- **CLI:** hidden `coder exp experiment-rules list|on|off|set|reset`.
- Write commands take `--expected-revision`. Without it they read the
current revision first, which only protects the gap between the two
calls; the help text says so.
  - A 409 prints the current rule and exits non-zero, with no retry.
- `reset` warns when the startup default is on, because reset is not a
kill switch.
- **Deployment Overview:** a note says the Experiments values are
startup defaults that runtime rules can override per user.
- **Docs:** a new "Target experiments at runtime" section in
`docs/reference/feature-stages.md` covers the modes, roles, CEL fields
and limits, reset vs. off, revisions and audit, the kill-switch
guarantee (including the rolling-upgrade requirement), frontend
freshness, fail-closed behavior, and that rules are not an authorization
boundary.

## Testing

- **HTTP:** every mode applies on the next request with no restart;
rejected writes keep the rule; RBAC (members 403 on GET and PUT;
auditors can read but not write; an invalid body from an unauthorized
caller still gets 403); malformed and ignored stored rows.
- **HA:** two coderd instances share one database. A PUT through A is
seen by B's `/api/v2/experiments` and by B's evaluator, which is the one
chatd uses.
- **Audit (enterprise, licensed):** an effective change creates one
entry whose diff carries the condition, and the entry is found before
the diff is checked. No-ops, 403s and 400s create no entry. A 409 and an
injected 500 are audited with an empty diff. A failed export keeps the
change.
- **Redaction:** a sentinel condition appears only in the rules API,
audit entries, and the server log's audit lines (Enterprise's slog audit
backend writes every audit diff there). It is absent from every other
log line, `/deployment/config`, the support bundle, and telemetry. The
test wires the slog audit backend so this is checked as stated.
- **CLI:** write and list, a conflict that is not retried, and the reset
warning.

Remote dogfood UAT through Coder Agents passed on a licensed test
deployment:
- the CLI;
- per-user results in the browser;
- 403 for members, including an invalid body;
- group targeting, including removing the user from the group;
- sentinel redaction across logs, `/deployment/config`, the support
bundle and the audit API (at that head the audit still masked the
condition; it is now tracked);
- the audit page showing the condition as changed but masked (same
note);
- the Overview note;
- the static default with `off` and `reset`;
- chatd enforcement toggled between turns.

That round tested `135aaaa41e`. The later commits are a CLI refactor and
a test addition, and a follow-up round will cover the final head.

## Follow-ups (low, not in this PR)

- A broken rule logs one error line per evaluation. Consider
rate-limiting it per experiment and revision.
- There's no API or CLI path to clear a stored rule for an experiment
that isn't user-scoped; those rules are inert and listed as `ignored`.
- A malformed stored rule is listed with zero `updated_by` and
`updated_at` values instead of omitting them.
- CLI wording: the 409 message, conflict output showing a UUID instead
of a username, a no-op write not saying nothing changed, and the parent
help lacking the `user.*` fields.
- `PutExperimentRuleRequest.mode` is not in the Swagger `required` list.
An omitted mode already returns 400.
- The dashboard's experiments query (from #29964) refreshes on mount,
window focus or reconnect once 60 seconds old, not on a timer, and its
comment says it bounds staleness. Correct the comment, or add a refetch
interval if the UI must pick up rule changes without a reload.

## Stack

1. #29869: evaluator library
2. #29898: storage and audit types
3. #29964: enforcement and discovery
4. **This PR:** control plane

> [!NOTE]
> This PR was written by an AI agent (Xum).

---
_Generated with [`xum`](https://github.com/coder/xum) • Model:
`anthropic:claude-opus-5-5` • Thinking: `high`_

---------

Signed-off-by: Thomas Kosiewski <tk@coder.com>
2026-09-29 12:23:00 +00:00
Sas Swart 813b8c151f feat: make AI Gateway content recording optional (#30031)
Add `--ai-gateway-disable-content-recording`, which stops user prompts,
tool
calls and model thoughts from being stored. Cost control reads none of
those
tables, so budgets, spend accounting and enforcement are unaffected.

A dropped record is still logged, so pairing the option with
`--ai-gateway-structured-logging-source=gateway` keeps the SIEM export
for
records that are never persisted. The gateway warns at startup when
content is
dropped while coderd is the only emitter, since those records then reach
neither
the database nor a SIEM.

Covered end to end for the embedded and the standalone gateway, and
documented
alongside cost controls.

<details>
<summary>Why the stack is ordered this way</summary>

The final design is a chain of recorder decorators:

    +---------------------+
    |    AsyncRecorder    |  all records except the interception start
    +---------------------+
    |     LogRecorder     |  emit the structured record
    +---------------------+
    | ValidatingRecorder  |  refuse malformed records
    +---------------------+
    |  caller middleware  |  record policy, e.g. drop content
    +---------------------+
    |    TraceRecorder    |
    +---------------------+
    |    DRPCRecorder     |  acquire client, send to coderd
    +---------------------+

Ordering constraints:

- The record type constants (1) are the vocabulary both emitters and the
  validator use, so they come first.
- Validation (5) rejects unset timestamps, so records must already be
stamped
where they are created (3). Stamping at creation leaves
`WrappedRecorder`
  doing nothing but client acquisition, which is why it is folded into
  `DRPCRecorder` (4) and not before.
- coderd's server-time fallback (2) is what keeps a gateway of an older
version
  from persisting zero times once the gateway owns stamping (3).
- PRs 1 to 5 change no user-visible surface. Each deployment option then
lands
with the code that reads it: the log source with gateway-side emission
(6),
content recording with the filter that drops the records (7). Pool
options are
plumbed to the serving pool in 6, where the first option that needs them
  arrives, rather than as an unused hook beforehand.

</details>

---
🤖 Raised by [Coder Agents](https://coder.com) on behalf of @SasSwart.
2026-09-29 11:48:53 +02:00
Sas Swart 23fb35d185 feat: let the AI Gateway emit the structured interception records (#30030)
coderd emitted the structured interception records as they arrived over
DRPC, so
a record the gateway never sends cannot be exported.

- Add `--ai-gateway-structured-logging-source` (`coderd`, `gateway`,
`both`) and
teach the gateway's `LogRecorder` to emit the same format where the
records
originate. Gateway lines omit `thread_parent_id` and `thread_root_id`,
which
  are resolved by a database lookup only coderd can perform.
- Reaching the serving pool required repairing the pool options: they
were
derived at construction sites and discarded, so the server always built
its
pool with the defaults. They are now derived from the deployment
configuration
in one place and passed to the pool that serves requests. The server
reports
those options, and its ownership of the `key_pool_state` collector, so a
second
  registration cannot silently reappear and panic startup.
- The standalone gateway inherits both structured logging options, since
it is
  now one of the processes that can emit.

This is the keystone of the stack: the flag, the mechanism that carries
it and
the behavior it selects land together, so
`TestAIGatewayStart_InheritedOptions`
and the gateway help output stay truthful at every commit.

<details>
<summary>Why the stack is ordered this way</summary>

The final design is a chain of recorder decorators:

    +---------------------+
    |    AsyncRecorder    |  all records except the interception start
    +---------------------+
    |     LogRecorder     |  emit the structured record
    +---------------------+
    | ValidatingRecorder  |  refuse malformed records
    +---------------------+
    |  caller middleware  |  record policy, e.g. drop content
    +---------------------+
    |    TraceRecorder    |
    +---------------------+
    |    DRPCRecorder     |  acquire client, send to coderd
    +---------------------+

Ordering constraints:

- The record type constants (1) are the vocabulary both emitters and the
  validator use, so they come first.
- Validation (5) rejects unset timestamps, so records must already be
stamped
where they are created (3). Stamping at creation leaves
`WrappedRecorder`
  doing nothing but client acquisition, which is why it is folded into
  `DRPCRecorder` (4) and not before.
- coderd's server-time fallback (2) is what keeps a gateway of an older
version
  from persisting zero times once the gateway owns stamping (3).
- PRs 1 to 5 change no user-visible surface. Each deployment option then
lands
with the code that reads it: the log source with gateway-side emission
(6),
content recording with the filter that drops the records (7). Pool
options are
plumbed to the serving pool in 6, where the first option that needs them
  arrives, rather than as an unused hook beforehand.

</details>

---
🤖 Raised by [Coder Agents](https://coder.com) on behalf of @SasSwart.
2026-09-29 11:48:52 +02:00
Seth ShelnuttandCian Johnston faa9815cd6 feat: configure Claude Platform for AWS on Anthropic providers (#29049)
This PR adds configuration plumbing for Claude Platform on AWS

## Summary

Makes Claude Platform for AWS configurable as a settings variant of
`type=anthropic`, discriminated by `_type: "claude_platform_aws"`.
Settings contain only `region` and `workspace_id`, plus the schema
version. No new provider type or database enum is introduced.

The API persists and projects those settings to the gateway. Provider
keys can be added or removed independently: allowed BYOK takes
precedence, followed by stored keys, then ambient SigV4 when neither key
source applies. Model-availability hints and keyless provider
construction reflect that ambient capability. Rejected or exhausted keys
do not fall back to IAM.

Removes Claude Platform-specific AWS identity fields, secret
redaction/merge behavior, external-ID handling, and mode/key consistency
checks. Bedrock configuration remains intact. Environment-based provider
seeding was removed upstream and is not restored. The Claude Platform
form and Experimental label remain follow-up work in #29050.

## Authentication limitation

Each Claude Platform provider uses either the gateway's ambient AWS
credentials when no provider keys are stored, or its stored API-key
pool. In either case, a client credential takes precedence when BYOK is
enabled. Rejected BYOK or an exhausted provider key pool does not fall
back to ambient IAM.

There is no per-provider AWS identity selection: keyless providers on
the same gateway share its ambient AWS credential chain and cannot
configure separate AWS access keys or assumed roles. Each provider still
has its own Claude Platform workspace ID and separate provider
attribution in Coder's usage/cost records; this does not imply separate
AWS billing accounts.

## Validation

- Race tests passed for SDK, daemon projection, db2sdk, chatprovider,
and targeted CLI/API configuration tests.
- Installed pre-commit hook passed, including generation, frontend/Go
lint, and build.
- Tests cover settings round trips, validation, adding/removing the last
key without settings changes, keyless construction, model availability,
and Bedrock regressions.

Generated by Coder Agents on behalf of @johnstcn and @Shelnutt2

Co-authored-by: Cian Johnston <cian@coder.com>
2026-09-29 09:20:31 +01:00
Seth ShelnuttandCian Johnston 10a22c8302 feat: support Claude Platform for AWS on the Anthropic provider (#29048)
## Summary

Adds Claude Platform for AWS as a variant of the Anthropic provider,
preserving the native Messages protocol and model IDs. Builds on the
merged SigV4 extraction in #29047. Configuration and UI follow in #29049
and #29050.

A provider-owned transport handles both bridged and passthrough
requests, sets the configured workspace header, and re-signs SDK
retries. Credential precedence is allowed BYOK, then the stored key
pool, then ambient AWS SigV4 when neither key source applies. Rejected
keys or an exhausted pool do not fall back to IAM. AWS configuration
loads lazily, so key-authenticated requests do not require working AWS
credentials.

There are no explicit authentication-mode, static AWS credential, Role
ARN, or external-ID fields for Claude Platform. Bedrock behavior and PRM
attribution remain separate. Anonymous Anthropic Messages support is out
of scope.

## Validation

- Full offline `go test -race ./aibridge/...` passed on this PR tip.
- Installed pre-commit hook passed, including generation, lint, and
build.
- Tests cover workspace-header signing, retries, key
precedence/failover, lazy credential recovery/cache, cancellation, and
stream lifetime.
- `TestClaudePlatformLive` sends a small real request through the bridge
when explicitly enabled. It was not run against AWS.

Co-authored-by: Cian Johnston <cian@coder.com>
2026-09-29 09:06:00 +01:00
Bobby Ho dffa890d8f ci: add macOS runner telemetry and a stall detector to test-go-pg (#30050)
Add telemetry to diagnose intermittent macOS `test-go-pg` stalls and preserve timeout diagnostics.

- Log runner details at startup; sample load, swap, and `vm_stat` every second and `top` every 10 seconds to identify runner-wide pauses.
- Add stall detection to `cli` and `agent` `TestMain`, logging heap, GC, and goroutine counts when ticks arrive more than 5 seconds late.
- Run artifact uploads with `always() && !success()` so canceled test steps do not skip diagnostics.
- Add a `TEST_TIMEOUT` override: use 20m on macOS to leave room for goroutine dumps within the 30m job, retaining the 25m default for the nightly gauntlet.

Validated with race tests, binary compilation, `actionlint`, local macOS checks, and Makefile timeout checks.
2026-09-29 03:41:18 +00:00
Jon Ayers 797dc6b2d1 feat: refresh header commands based on JWT expiration (#29760)
Adds opportunistic JWT parsing for values returned from the
`--header-command`. If a JWT is parsed we use the value of its `exp`
field to determine when we should re-run the command to refresh the
values.
2026-09-28 09:42:31 -05:00
Michael Suchacz f5092aaa4a feat: make chat limits configurable deployment options (#28927)
Adds deployment options for the chat limits that shape turn behavior and
API acceptance, replacing hardcoded constants in `coderd/x/chatd` and
the chat API. Defaults are unchanged.

| Option | Default | Limit |
|---|---|---|
| `CODER_CHAT_MAX_STEPS_PER_TURN` | 1200 | steps (model responses) in a
chat turn |
| `CODER_CHAT_MAX_GENERATION_RETRIES` | 25 | consecutive retries after a
model generation fails with a transient error |
| `CODER_CHAT_MAX_QUEUED_MESSAGES_PER_CHAT` | 20 | messages that can be
queued in a chat |
| `CODER_CHAT_MAX_ATTACHMENTS_PER_CHAT` | 50 | files linked to a chat;
linking more deletes the earliest uploads |
| `CODER_CHAT_MAX_PROMPT_BYTES` | 131072 | system prompt, plan mode
instructions, and each user's custom prompt |
| `CODER_CHAT_MAX_CONCURRENT_RECORDING_UPLOADS` | 25 | desktop
recordings each server stores at the same time |

`codersdk.DefaultChat*` constants remain the single source of truth.
Values below 1 fail startup validation, and each option's description
ends with "Must be at least 1." `chatd.New` fills any zero field in
`chatd.Config.Limits` with its default once, `LimitsFromConfig` is a
plain conversion, and the code that receives a limit as a parameter (the
`chatstate` queue and file link checks, the turn step check,
`chatretry.Retry`, and `chatadvisor.NewRuntime`) returns an error for a
limit below 1 instead of choosing a default or disabling the limit. The
options are documented in the generated configuration reference and in
the platform controls docs.

`CODER_CHAT_MAX_GENERATION_RETRIES` counts consecutive retries, and the
count resets after each successful step. The default allows 25 retries
(26 attempts) where the old constants allowed 24. The same limit now
applies to each nested advisor model call and to background title,
summary, and turn status label generation, which previously used fixed
`chatretry` budgets.

Chat creation and tool result bodies keep the fixed 256 KiB limit from
main. The prompt-saving endpoints accept bodies up to twice the prompt
limit, for JSON escaping and characters that sanitization removes, but
never less than 256 KiB, so small prompt limits still return the 400
length error. At the default this equals the 256 KiB limit on main.

Four settings from earlier revisions were removed, and the fixed limits
from main are back:

- `CODER_CHAT_MAX_TOOL_OUTPUT_BYTES`: the agent already keeps a 16 KiB
head and a 16 KiB tail of command output, so raising it had no effect
and lowering it cut the tail.
- `CODER_CHAT_MAX_DYNAMIC_TOOLS_PER_CHAT`: tool definitions share the
256 KiB chat creation body limit, so raising the count did little.
- `CODER_CHAT_DEBUG_MAX_BODY_BYTES`: one setting covered two unrelated
debug caps.
- `CODER_CHAT_DEBUG_MAX_TEXT_RUNES`: the name did not say what it
applied to.

Root and subagent concurrency caps stay fixed because they are licensing
behavior, and `MaxChatFileSizeBytes` stays fixed because the web UI
hardcodes the generated constant for client-side checks.
`codersdk.MaxChatFileIDs` is removed; the web UI's "Attachment expired"
tooltip no longer quotes the number because the cap is now per
deployment, and it says that attaching a file beyond the limit
permanently deletes the chat's earliest-uploaded attachments.

## Review record

| Round | Reviewed head | Result | Disposition |
|---|---|---|---|
| Codex 1 and 2 (Sep 2 to 3) | `7f10a1572f8`, `8ee09f8fa3b`,
`197bb78d5b` | 10 findings, then OK on `197bb78d5b` | all threads
resolved |
| Codex 3 (Sep 17) | `e4efac249a5` (pre-rebase) | 4 findings | advisor
retry limit confirmed and fixed in `9e462b7b3d0` (rebased from
`67341e458e8`); ARCHITECTURE.md TODOs added; JSON-escaping body sizing
and streamed-argument rune cap declined as pre-existing design (evidence
in threads) |
| Codex 4 (Sep 18) | `9e462b7b3d0` | 6 findings | recording evicted at
attachment cap 1 confirmed and fixed in `5b61895decb`; `Limits` field
docs and zero-value fallback documented; prompt retro-enforcement,
generated multi-sentence lines, and tool-output truncation metadata
declined as pre-existing design (evidence in threads) |
| Codex 5 (Sep 18) | `5b61895decb` | 1 finding | over-cap batch of
already-linked files accepted after lowering the cap: confirmed and
fixed in `f44063e74e5` |
| mafredri (Sep 23) | `f44063e74e5` | 24 threads, changes requested |
all fixed in `05e3e3d622a`, `c1f78e09e8b`, `709ddb67977`, `d0587a49829`,
`ca4347b24de`, and `1f6c9e2a46f`, then answered and resolved; the
ARCHITECTURE.md text was written at the author's direction and approved
by the author |
| Sibling sweep (Sep 24) | `1f6c9e2a46f` | author-requested audit of all
24 mafredri threads for sibling issues, plus a test refactor | fixed in
`37eaf07d34d` through `1f14ac041b7`: the advisor retry limit,
`chatretry.Retry`, and the turn step check reject values below 1 instead
of disabling retries or the limit; limit tests run against both the
default and configured values set through `chatd.Config` or
`DeploymentValues`; the expired-attachment tooltip is concrete. A
required coder-frontend-review run (8 roles in one agent, CD re-run
once) found a stale story comment, removed. |
| Codex 6 (Sep 27) | `a71383e3a31` | OK, no findings | requested after
the rebase at the author's direction, past the six-pass review budget |

The branch was rebased onto main `4851c7b60c8` on Sep 27; the previous
head was `91b84000adb`, and the earlier SHAs in the review record
predate the rebase. Conflicts in the `chatstate` transitions and
`CreateChat` keep both main's `InitialStatus` and this PR's
`MaxFileLinks`, and main's rewrite of `advisor_internal_test.go` in
#28754 keeps main's version plus this PR's limit tests. Two tests added
on main needed the new retry arguments:
`TestGenerateStructuredTitleWithUsage_FallsBackToTextWhenToolChoiceRejected`
and `TestAdvisorRunStreamsReasoningDeltasSeparatelyFromAdvice`.
Validation on `a71383e3a31`: the pre-commit hook (generation,
golangci-lint, build), the full `coderd/x/chatd/...` suite (20
packages), the chat tests in `codersdk`, `coderd`, and
`enterprise/coderd`, the CLI help tests, race-detector runs of the
`chatd`, `chatstate`, `chatadvisor`, and `chatretry` packages, and `go
build` plus `go vet` on each of the 27 commits. No remote UAT was run.

> Xum acted on behalf of @ibetitsmike for this pull request.
2026-09-28 09:25:01 +00:00
Michael Suchacz 6826da9397 feat: restore agents-access as a default organization role (#30004)
Restores `agents-access` (**Coder Agents User**) as a built-in
organization role and adds it to every organization's
`default_org_member_roles`. Members keep Coder Agents by default, and
admins can now restrict it per organization by removing the role from
the defaults and granting it through IdP role sync or direct assignment.

Since #28186, chat permissions lived in the organization-member floor,
so the only way to limit Coder Agents was narrowing per-model access
lists, which did not hide the feature.

- Migration 000602 appends `agents-access` to every organization's
defaults, so upgrading keeps every member's access. New organizations
get it from `rbac.DefaultOrgMemberRoles()`.
- Service accounts do not inherit the role from the defaults
(`GetAuthorizationUserRoles`), so they still need an explicit grant for
chat.
- Owners and Organization Admins keep chat access regardless of
defaults. Chats shared with a user stay readable without the role.
- Role sync is unchanged. It does not store default roles, so after the
default is removed a mapped user gets `agents-access` at their next
sign-in.
- The retired-role mechanism (`IsRetiredRoleName`) is removed.
- `PATCH /api/v2/organizations/{organization}` rejects duplicate
`default_org_member_roles` entries (the CLI already de-duplicated them).
- Custom organization role requests treat built-in role names
case-insensitively, since custom role names are stored lowercase and
`Agents-Access` would otherwise become an undeletable shadow of the
built-in role.
- The Agents permission alert tells restricted members their chats are
kept, and the model selector for users without chat permission no longer
claims that no models are configured.

Docs add "Control who can use Coder Agents" to the Agents
getting-started guide.

**NOTE**: Migration 000602 deletes custom roles named `agents-access`,
which the built-in role would shadow. The down migration does not
restore them. Stored member grants of `agents-access` from the earlier
built-in role or a deleted custom role stay and now grant chat access.

## Validation

- Remote dogfood
[UAT](https://dogfood.cdr.dev/agents/e1c18dc6-f6f0-49a1-b6cd-6ce014b51a86)
passed on fb441dfae5, covering case variants of built-in role names on
create, update, delete, default roles, and member grants. The previous
round, on a6fef66b3f, passed the upgrade from main 5dde49dffb on a
fresh database. 6b7f107e0b only changes wording (docs, comments, UI
copy, CLI help) and did not go through UAT.
- Local: rbac, dbauthz, idpsync, migrations (chain including 000602 down
and up), querier, httpmw, provisionerdserver, coderd and enterprise
chat, role, and organization tests, chatd, toolsdk, CLI; frontend
vitest, story tests, biome, oxlint, tsc.

## Known issues (not changed here)

- Sending from a chat page that was already open when access was removed
shows a generic 404 panel until reload.
- The members page role editor is not locked for OIDC users under org
role sync; the API rejects the change with 400.
- Creating a custom role named after a system role
(`organization-member`, `organization-service-account`) returns 400 with
a raw database unique-constraint error.

<details>
<summary>Review record</summary>

- Remote UAT (UAT Critic sub-agent `6636c855ec`): round 1 on 8bec029d35
passed with issues, fixed in d7ba937a10 and e0f16e1a45f; round 2 on
a6fef66b3f passed; round 3 on fb441dfae5 passed.
- Codex: round 1 on a6fef66b3f found one P2 (case variants of built-in
role names), fixed in fb441dfae5 with a red-green test. Rounds 2 and 3
on fb441dfae5 found no major issues.
- mafredri on fb441dfae5 (PR text only): nine wording suggestions
applied in 6b7f107e0b9; NOTE extended for stored grants; description
trimmed.
- Frontend: FE1 to FE10 and coder-frontend-review checkpoints
self-audited.

</details>

> 🤖 This PR was opened by [Xum](https://mux.coder.com) on behalf of
@ibetitsmike.
2026-09-27 23:34:26 +02:00
Ethan f64ccfb836 feat: add experimental inline MCP servers (#29546)
Closes CODAGT-1131

Fifth PR in the inline MCP servers stack, on top of #29520. A chat owner
can now declare MCP servers by URL and headers on their own chat, and
chatd connects to them every turn alongside the org-configured ones.

They go in a new `inline_mcp_servers` field on `POST /api/v2/chats` and
`POST /api/v2/chats/{chat}/messages`, which replaces the whole set (omit
to keep, `[]` to remove). Root chats only. `GET /api/v2/chats/{chat}`
returns them, with the URL blanked for non-owners and header values
never returned. Subagents only get servers marked `allow_in_subagents`,
a root chat in explore mode gets none, and a root chat in plan mode gets
all of them.

URLs must be `https`, or `http` to an IP literal in
`CODER_MCP_ALLOWED_PRIVATE_CIDRS`. Header values have no minimum length,
so anything under the redactor's 8 byte floor from #29518 is accepted
but not redacted, which the docs call out. Every cap is exported from
`codersdk`.

It's behind the `chat-inline-mcp-servers` experiment, and the
`--disable-chat-caller-supplied-tools` kill switch from #29466 covers it
too. With either off, declaring servers gets a 403 and existing chats
stop connecting to theirs, but `[]` still works so a client can detach.

Load failures show up in the chat debug panel as failed connections, as
do org servers with an invalid config. Docs are hand-written in
`mcp-servers.md`.
2026-09-25 18:51:55 +10:00
Ethan c1af0793e5 feat: add caller-supplied tools kill switch for chats (#29466)
This is the first PR in a stack that adds experimental support for
attaching MCP servers to a chat when it's created, and then deprecates
`unsafe_dynamic_tools` in favour of them. Both are "caller-supplied
tools", i.e. tools the client hands to coderd at chat creation rather
than tools an admin has configured, so we want a single switch admins
can use to turn all of them off.

This PR adds that switch: `--disable-chat-caller-supplied-tools`
(`CODER_DISABLE_CHAT_CALLER_SUPPLIED_TOOLS`). It follows the existing
`disable-*` deployment flag pattern (`--disable-chat-sharing` etc.), so
there's no new API route and it's independent of any experiment. When
set, `POST /api/v2/chats` returns a `403` if the request includes
`unsafe_dynamic_tools`, and chats that already have dynamic tools run
without them; a call to one gets a tool error result and the model just
continues. Later in the stack, chat-attached MCP servers are gated
behind the same flag.

It's unset by default, so nothing changes unless you opt in.


Part of CODAGT-1131
2026-09-25 18:51:53 +10:00
Nick Vigilante 99a3425333 fix: advertise install.sh with curl -fsSL everywhere (#29801)
## Summary

The install one-liner is advertised as `curl -L` in most docs pages, the
README, and the CLI's default version-mismatch hint, while `install.sh`
itself and the deployment-local hint use `curl -fsSL`.
Without `-f`, curl writes an HTTP error body to stdout, and the
one-liner pipes that error page into `sh`.
This PR standardizes every `install.sh` invocation on `-fsSL` so a
failed download fails closed.

## Changes

- Docs: `install/cli.md`, `install/operate/upgrade.md` (twice),
`get-started/index.md`, `start/local-deploy.md`,
`install/operate/provisioners/index.md`,
`admin/templates/managing-templates/change-management.md` (twice),
`user-guides/workspace-access/zed.md`, and `README.md`.
- CLI: `defaultUpgradeMessage` in `cli/root.go` now prints `curl -fsSL`,
matching the deployment-local branch in the same function.
- `codersdk/deployment.go`: the `--cli-upgrade-message` description now
says `curl -fsSL`; the server help goldens, `server-config.yaml.golden`,
`docs/reference/cli/server.md`, and
`docs/admin/setup/configuration-reference.md` are regenerated to match.
- `install.sh`: the "To run" header comment now matches the script's own
usage text.
- Style guide: a new "Commands that pipe a download into a shell" rule
in `docs/.style/style-guide/formatting.md`, and the tabs example uses
`-fsSL`.

The coder.com download page already builds its command from `-fsSL` in
coder/coder.com#1093, and coder.com `main` has no other `install.sh`
one-liners outside test fixtures, so no coder.com change is needed.
DatoCMS content and a release verification page are not in this PR.

`main` since moved the upgrade and provisioner pages under
`docs/install/operate/` and removed the install one-liner from
`docs/install/index.md` in #29304, so that page no longer appears in
this diff.

## Testing

- `TZ=UTC go test ./cli
-run='Test(CommandHelp|ServerYAML|ErrorExamples|.*Golden)' -update` and
`go test ./enterprise/cli -run=TestEnterpriseCommandHelp -update` pass
and produce the committed golden diffs.
- `go test ./cli -run 'Test.*Version|TestWrapTransport'` passes,
including `Test_wrapTransportWithVersionCheck`.
- `go run ./scripts/configdocgen` output contains the new `curl -fsSL`
description.
- `markdownlint-cli2` on the edited style guide page reports 0 errors.
- `git grep` finds no remaining `curl -L …install.sh` outside the style
guide's **Don't** example.
- After merging `main`: `make pre-commit` passes, and `make test`, `make
test-js`, and `make site/out/index.html` pass.

Linear:
https://linear.app/codercom/issue/DOCS-964/standardize-installsh-curl-flags-on-fssl-across-docs-codercom-and-the

---
🤖 Built with AI assistance.
2026-09-24 18:23:54 +00:00
陈志谦 442b08fee5 docs: drop duplicated 'the the' in scaletest flag help and OIDC docs (#29107)
Two doubled-word typos:

- `cli/exp_scaletest.go`: the `--connect-url` flag description shown by
`coder exp scaletest --help` read "inside **the the** workspace over
WireGuard"
- `docs/admin/users/oidc-auth/index.md`: "as **the the** username"

Docs/help text only.
2026-09-24 16:19:53 +00:00
Spike Curtis 9c0c3948c9 feat: narrow ssh and bash tool client.Workspace calls with include_related (#29542)
Part of the lite `codersdk.Workspace` work
([GRU-82](https://linear.app/codercom/issue/GRU-82),
[RFC](https://app.notion.com/p/coderhq/Lite-Workspace-API-Requests-3aad579be5928060b5e7cc65c539717f)).
Stacked on #29541.

Narrows the two production `client.Workspace` (get-by-ID) call sites
that read only a subset of the returned workspace.

- **`cli/ssh.go`** (`notifyCondition`) reads `TTLMillis` and
`LatestBuild.Deadline`. `TTLMillis` is computed server-side from the
template (`AllowUserAutostop`/`DefaultTTL`), so this requests `template`
and `latest_build` (the deadline is on the build itself; no job,
resources, or template version needed).
- **`codersdk/toolsdk/bash.go`** (`findWorkspaceAndAgent`) refetches
after autostart only to locate the first agent, so it requests
`latest_build.resources.agents`. Callers read only the agent plus the
top-level `Name`/`OwnerName`.

The other production by-ID callers were left on the default (full)
selection because they need the whole object: `support/support.go`
serializes the workspace into the support bundle,
`codersdk/toolsdk/chatgpt.go` JSON-marshals it for the model, and
`codersdk.ResolveWorkspace` is a generic resolver whose callers expect
everything. CLI paths that use `WorkspaceByOwnerAndName` are not touched
because that endpoint does not parse `include_related` yet.

### Testing

- `TestWorkspaceBash` and `TestWorkspaceBashTimeout` pass; the
autostart-refresh branch has no dedicated test, so its selection was
verified by tracing the fields read from the returned workspace and
agent.
- `cli` and `codersdk/toolsdk` build/vet/lint clean; the SSH deadline
notifier has no dedicated test, so its selection was verified the same
way (and by confirming the server derives `TTLMillis` from the
template).

---

*Generated by Coder Agents on behalf of @spikecurtis.*
2026-09-24 16:34:08 +02:00
McKayla はな 0c98308ca9 feat: default the CLI workspace search to the user filter (#29753) 2026-09-23 19:42:10 -06:00
Callum Styan 83bef8a12d fix: consolidate pg and nats pubsub metrics generation into one (#26843) 2026-09-22 14:00:44 -07:00
Michael Suchacz 988520477e feat!: remove the experimental chat API mounts after the transition window (#29548)
Remove the `/api/experimental` duplicates of the promoted chat API.
#28496 double-mounted the chat, organization chat and MCP server, member
model-override, AI provider key, and MCP OAuth2 disconnect routes under
both prefixes for the one-release compatibility window decided in
CODAGT-921; that window closed with v2.37.0, so these routes now answer
only on `/api/v2`. The change also deletes what existed only for the
dual mount: the `chatAPIPrefix` plumbing, the shared cross-prefix files
rate limiter (`RateLimitByAPICompatibilityEndpoint`), the 410 tombstones
for the legacy `chats/providers` and `chats/user-provider-configs`
collections with their `ExperimentalClient` SDK methods, and the
`/providers` and `/user-provider-configs` 404 reservations on `/api/v2`.
Signed chat file download URLs now point at `/api/v2`.

Routes that were never promoted stay on `/api/experimental`: the
experiment-gated `chats/config/advisor` and
`chats/config/computer-use-provider`, `chats/{chat}/stream/desktop`,
`chats/{chat}/debug/runs`, and the MCP OAuth2 callback, whose path is
pinned as a registered redirect URI with external providers. The
frontend already calls `/api/v2` for every promoted route, so only
`typesGenerated.ts` changes. Because the experimental mount no longer
reserves the removed collection segments, requests such as `GET
/api/experimental/chats/models` now fall into the `{chat}` wildcard and
return 400 instead of 404; the experimental prefix carries no stability
guarantee and the compatibility window is over, so no new reservations
were added.

`coderd/x/chatd/ARCHITECTURE.md` now describes the single `/api/v2`
mount and the remaining experimental routes, and its endpoint headings
use the `/api/v2` paths.

Closes CODAGT-922
(https://linear.app/codercom/issue/CODAGT-922/remove-experimental-chats-api-paths-after-transition-window).
Refs #28496.

## Delivery record

- Remote dogfood UAT (Coder Agents chat
https://dogfood.cdr.dev/agents/c016b549-201a-4eca-a50f-3c0b7c339db6):
verdict PASS on b008976a4dfba3524b220ca846ea3c3492e9cc9c. The branch was
since rebased onto main (which had a duplicate migration 000596 until
#29568) and gained one docs-only commit
(`coderd/x/chatd/ARCHITECTURE.md` only); the current head 666332f579
carries the same code diff, so the tested branch code is unchanged.
Covered the Agents UI end to end on `/api/v2` (chat, streaming reply,
reload mid-stream, file upload and download, debug runs panel), the
admin AI settings pages with persisted saves, 45 curl route checks
across both prefixes, websocket handshakes, the frozen MCP callback
redirect URI, and the files rate limiter. One low-severity note: the
400-versus-404 behavior for removed experimental collection paths
described above.
- Codex review on b008976a4d (draft marked ready): no findings. Codex
review on the current head 666332f579 (requested via `@codex review`):
one P1 thread asking for TODO markers instead of the direct
`ARCHITECTURE.md` edit; replied and resolved because the PR author
reviewed the earlier TODO markers and directed this replacement text. No
unresolved threads. CI is green on 666332f579 (run 35376339802;
`test-go-pg-17` passed on rerun after a Docker Hub timeout in
`TestSSH_Container/OK`).

> [!NOTE]
> Xum acted on @ibetitsmike's behalf in this pull request.
2026-09-21 17:42:31 +02:00
Ehab Younes 9974294ca9 feat: report arbitrary workspace agent session types (#28338)
Report session counts as a dynamic app-name map from the agent, the SSH CLI, and the workspace usage API, so a new IDE flows end to end without a schema, proto, or server change.

- Report through `Stats.session_counts` on Agent API v2.11. Older agents keep reporting through the deprecated count fields.
- Accept arbitrary names from `coder ssh --usage-app` and the workspace usage API. A name the registry does not recognize keeps its identity and is attributed to the `unknown` family.
- Add reconnecting PTY counts to the map without overwriting the counts SSH reported.
- Normalize and merge names before applying the deterministic 64-name cap, ordered by known apps, then count, then name. Overflow sums into `unknown`, and non-positive counts cannot cancel valid activity.
- Strip control characters, treat whitespace-only HTTP app names as absent, and require the exact environment variable prefix when you extract an SSH session type.
- Count overflow entries in `coderd_agentstats_session_counts_overflow_total` and keep raw payloads out of logs.
- Change the usage API's `app_name` to a string, and update the SDK types and API reference.

closes #27410
2026-09-19 00:32:51 +02:00
Callum Styan 057475c4e8 feat: control notification scaletest load via template deletion count (#29151) 2026-09-18 11:57:06 -07:00