Commit Graph
89 Commits
Author SHA1 Message Date
Garrett Delfosse 7394a66e72 test(codersdk/toolsdk): fix TestTools/WorkspaceExecute workdir check on Windows (#32323)
Fixes CODAGT-1333.

`TestTools/WorkspaceExecute` failed on Windows with `GetFileAttributesEx
/tmp: The system cannot find the file specified.` The agent runs
commands via `sh -c` on every platform (`agent/agentproc/process.go`),
which is Git Bash/MSYS on Windows runners. There `pwd` prints a POSIX
path like `/tmp/...`, and `filepath.EvalSymlinks` cannot resolve it. The
command itself ran in the right directory; the assertion was wrong.

The test now runs `touch marker.txt` in the requested workdir and
asserts the file exists at `filepath.Join(workDir, "marker.txt")` on the
host. This is the same approach `agent/agentproc` `CustomWorkDir`
already uses for this reason. No timeouts changed.

---
_Generated by Coder Agents on behalf of @f0ssel._
2026-11-17 09:50:43 -05:00
John T Skarbek 54fe847c6f chore: remove references to torn-down dev.coder.com instance (#31702)
The `dev.coder.com` instance has been torn down and now only redirects
to another instance. Nothing in the repo calls it. Every reference was a
test fixture, mock, story, Helm test input, docs example, or comment.

Replace all occurrences with `coder.example.com` and update the comments
that claimed the mocks were fetched from that instance. The Helm golden
files were updated to match their inputs.

Closes DOG-43

https://linear.app/codercom/issue/DOG-43/scan-codercoder-repo-for-remaining-references-to-devcodercom-and
2026-11-13 16:38:44 +00:00
Ethan a40ff12c8c feat!: steer chat messages by default and remove the interrupt busy behavior (#31174)
Breaking API Change for changelog:

> `POST /api/v2/chats/{chat}/messages` now steers by default. If the
chat is busy, the message is delivered before the agent's next model
call, instead of after the turn ends. `busy_behavior` accepts `steer` or
`queue`, and `interrupt` now returns a 400. To redirect the agent, just
send a message, and only call `POST /api/v2/chats/{chat}/interrupt` to
stop it. Queued messages now include their `busy_behavior`.

This is where steering becomes the default for API clients. A send
without a `busy_behavior` used to wait for the turn to end, and now it
goes in the steer lane, so the agent sees it before its next model call
and keeps going. `queue` still waits for the turn to end.

Interrupt isn't a busy behaviour any more. Since #31155 it just meant
"interrupt, then steer", so the API now returns a 400 for it, and
`codersdk.ChatBusyBehaviorInterrupt` is gone. `coder_send_chat_message`
in the MCP server gets the same default and enum. A steer message
already corrects a busy agent, and combining it with an interrupt goes
wrong in either order, so the docs and the MCP tool descriptions now say
to send a message to redirect the agent, and to interrupt only to stop
it.

Queued messages now report their `busy_behavior`, and the queue lists
the steer lane first, so clients can show which messages go in at the
next step and which wait for the turn to end. A message moved with Send
now reports `steer`.

The site doesn't send a `busy_behavior`, so its sends start steering
with this PR, but its queue UI doesn't know about the steer lane until
#31173. chatd still accepts `interrupt` internally, as `message_agent`
uses it, and #31175 removes both.

Part of CODAGT-1140
2026-11-09 15:55:17 +11:00
Kyle Carberry 36b472991d refactor: replace aisdk-go tool types with codersdk/toolsdk/tooldef (#31234) 2026-11-05 09:45:21 -05:00
coder[bot]andNick Vigilante a7f3e3b685 docs: promote the AI sections and Templates to the top level of the sidebar (#31645)
> [!NOTE]
> Generated by Coder Agents. The branch and this description were
written by an AI agent and need human review.

## Summary

First of three stacked PRs for the docs IA overhaul (Proposal 1, section
3). This PR dissolves `docs/ai-coder/` into four top-level sidebar
sections, in this order, where "Run AI coding agents in Coder" used to
be: **Coder Agents**, **AI Governance** (with Agent Firewall and AI
Gateway nested under it), **Agent Relay**, and **MCP server**.

- Files move with `git mv`, so `git log --follow` keeps their history.
The move includes every file in the moved folders, including pages that
are not in the sidebar, such as `agent-firewall/example-policy.md`, and
pages added to `main` after the first version of this PR, such as
`agents/projects.md` (#30684) and `agents/acp-subagents.md` (#31430).
- `docs/ai-coder/index.md`, `best-practices.md`, and `ide-agents.md` are
deleted. Each has a redirect (see below).
- `docs/manifest/` mirrors the new folders: `coder-agents.yml` (+
`coder-agents/`), `ai-governance.yml` (+ `ai-governance/`),
`agent-relay.yml`, and `mcp-server.yml`. `ai-coder.yml` is deleted. The
Projects and ACP subagents entries from `main` move with Coder Agents.
- Every relative link and image path is fixed, both in the moved pages
(their base directory changed) and in pages that link to them. This was
done in one scripted pass from a clean `main` checkout. Every link was
then checked to resolve on disk.
- Commit 2, `chore: point in-product docs links at the new docs paths`,
updates docs URLs outside `docs/`: `site/` `docs()` links, Go error
strings and comments, `README.md`, `server.json`, templatebuilder
modules, and the doc-check skill. Commit 4 reverts the unversioned
`https://coder.com/docs/...` URLs among them (see below), so after this
PR only these differ from `main` outside `docs/`: the `site/` `docs()`
links, the versioned `cli/cliui` troubleshooting links, and repo file
paths in comments and the doc-check skill's examples
(`docs/coder-agents/...`, `docs/templates/...`). You can drop commit 2;
the redirects cover the old URLs.
- **#31646 (Templates) was squash-merged into this branch** as
`98377a3a3b`. This PR now also carries the Templates move
(`docs/admin/templates/**` → `docs/templates/**`, plus
`template-from-scratch.md` and `testing-templates.md`) and its chore
changes. Refer to #31646 for that part's move map, redirects, and
review.
- Commit 4, `fix: keep docs links healthy before the release`, keeps
links working on the live site until a release that includes this stack
becomes the default docs version:
- Hand-written pages that linked to moved pages with absolute
`https://coder.com/docs/...` URLs now use relative links, anchors kept.
The new paths 404 on the live site today, but relative links resolve for
every version and are checked locally.
- Hardcoded, unversioned `https://coder.com/docs/<new path>` URLs
outside `docs/` that the chore commits introduced (Go strings including
the prebuilds metric help, CLI help, the
`agentFallbackTroubleshootingURL` default, `README.md`, `server.json`,
examples, templatebuilder files, and one `site/` constant) go back to
their current paths. Then `make gen` regenerated the generated table in
`docs/admin/integrations/prometheus.md`,
`scripts/metricsdocgen/generated_metrics`, the CLI golden files, and
`docs/reference/cli/templates/{create,edit}.md`. See "Post-release
follow-up".
- The `site/` `docs()` changes and the `cli/cliui` links stay, because
those URLs are versioned (`/docs/@vX.Y.Z/...`) and resolve once the
release ships.
- Commit 5, `docs: prefix relative links in the moved pages with ./`,
adds `./` to 13 relative links that the move script wrote without it, in
`docs/mcp-server.md` and three Templates pages.
They resolved, but every other relative link in the docs starts with
`./` or `../`.
- Commit 6, `docs: keep the coding agent module guidance in the
templates docs`, adds an **AI agent modules** section to
`docs/templates/extending-templates/modules.md`.
Deleting `docs/ai-coder/index.md` removed the only instructions for
adding a terminal coding agent to a template with a registry module.
The example (`registry.coder.com/coder/claude-code/coder`, `version =
"~> 5.2"`) passes `terraform validate`, and the Template Builder labels
it cites match `site/src/pages/TemplateBuilder/ModuleSelectStep.tsx`.

Stack: **this PR** (with #31646 squash-merged in) →
`docs/ia-tutorials-nav` (#31647, Tutorials + nav renames) →
`docs/ia-workspaces-folder` (#31650, folder rename).

Refs: Linear DOCS-1154, DOCS-1179, and DOCS-1479 for the coder.com and
router redirect fix.

## Move map

| Old | New |
| --- | --- |
| `docs/ai-coder/agents/**` (25 files) | `docs/coder-agents/**` |
| `docs/ai-coder/ai-governance.md` | `docs/ai-governance/index.md` |
| `docs/ai-coder/agent-firewall/**` (9 files) |
`docs/ai-governance/agent-firewall/**` |
| `docs/ai-coder/ai-gateway/**` (26 files) |
`docs/ai-governance/ai-gateway/**` |
| `docs/ai-coder/agent-relay/**` (3 files) | `docs/agent-relay/**` |
| `docs/ai-coder/mcp-server.md` | `docs/mcp-server.md` |
| `docs/ai-coder/index.md` | deleted |
| `docs/ai-coder/best-practices.md` | deleted |
| `docs/ai-coder/ide-agents.md` | deleted |

## Redirects added

Appended to `docs/redirects.json`, using the absolute `/docs/...` format
that `main` validates today. Wildcards cover every page under each moved
folder, including pages that are not in the sidebar and the pages added
to `main` since the first version of this PR.

| Source | Destination |
| --- | --- |
| `/docs/ai-coder` | `/docs/coder-agents` (deleted hub, **please
confirm**) |
| `/docs/ai-coder/best-practices` | `/docs/coder-agents` (deleted,
**please confirm**) |
| `/docs/ai-coder/ide-agents` | `/docs/user-guides/workspace-access`
(deleted, **please confirm**) |
| `/docs/ai-coder/ai-governance` | `/docs/ai-governance` |
| `/docs/ai-coder/mcp-server` | `/docs/mcp-server` |
| `/docs/ai-coder/agents/:path*` | `/docs/coder-agents/:path*` |
| `/docs/ai-coder/agent-firewall/:path*` |
`/docs/ai-governance/agent-firewall/:path*` |
| `/docs/ai-coder/ai-gateway/:path*` |
`/docs/ai-governance/ai-gateway/:path*` |
| `/docs/ai-coder/agent-relay/:path*` | `/docs/agent-relay/:path*` |

None of the 3 existing rules in `docs/redirects.json` points into
`ai-coder/`, so no destinations needed updating.

**coder.com and router static redirects.**
coder.com applies its own `redirects.json` and the rules it generates
from this file before it hands `/docs` to the docs-engine router, which
applies its own static `redirects.json` before the page.
Three older static rules in both repos send `/docs/templates`,
`/docs/templates/open-in-coder`, and `/docs/templates/troubleshooting`
to `/docs/admin/templates/...`, and the `/docs/admin/templates/:path*`
rule here sends them back, so each loops once a release with this layout
is the docs mainline.
A static `/docs/workspaces` rule would also hide the new
`docs/workspaces/index.md` page from #31650.
coder/coder.com#1201 and coder/docs-engine#304 (draft, DOCS-1479) remove
those rules; they must merge when that release becomes the mainline, not
before, because the old-layout mainline still needs them.
A chain simulation of every page and static source on each branch of
this stack finds no other loops or hidden pages.
The `/docs/ai-coder/...` and other legacy rules there still work; some
now take two redirects.

## Deviations from the proposal

- **Glossary "IDE agent" entry.** The proposal says to remove this link
and leave plain text. I pointed it at Access your workspace
(`docs/user-guides/workspace-access/index.md`) instead, the same page
the `ide-agents` redirect uses, because "Refer to IDE agents." with no
link reads as a dead end. The "Coding agent" entry now links to Coder
Agents, as the proposal says.
- **Top-level icons.** The new sections need icons. I chose `wand.svg`
(Coder Agents, the old ai-coder icon), `ai_intelligence.svg` (AI
Governance), `link.svg` (Agent Relay), and `plug.svg` (MCP server).
- **Coding agent module guidance.** Kept on the Reuse template code page
(commit 6) instead of deleted with `docs/ai-coder/index.md`. The
best-practices and IDE-agent pages are deleted as proposed.
- **Order relative to Administration.** In this PR the four sections sit
where `ai-coder` was, after Administration. PR C moves Administration
after MCP server to reach the final order.
- **`docs/mcp-server.md` repo link.** A relative link to
`codersdk/toolsdk/toolsdk.go`, outside `docs/`, was re-based for the new
depth.
- **Left unchanged on purpose:**
- Database migrations and the notification `.golden` files rendered from
them still contain `coder.com/docs/ai-coder/...` URLs. Existing
migrations can't be edited, so those links rely on the redirects.
- `aibridge/client/client.go` links to a commit-pinned GitHub permalink,
which still resolves.
- Test fixtures in `scripts/docsredirectscheck` and
`.github/workflows/test-docs-preview-mapper.sh` use `ai-coder` as sample
data.
-
`site/src/pages/AgentsPage/components/ChatElements/tools/Tool.stories.tsx`
uses `https://coder.com/docs/ai-coder/ai-gateway` as a sample web search
result. Changing it would only change a visual snapshot.

## Open questions for Nick

1. Do you agree with the three redirect targets for the deleted pages
above?
2. Should the "IDE agent" glossary entry link to Access your workspace
(current) or be plain text (proposal)?
3. Are the icons for the new top-level sections OK?
4. Open PRs that touch the moved files, checked with `git merge-tree`
against this branch. I didn't change them.
- #31169 merges cleanly: Git's rename detection carries its edit to
`docs/ai-governance/ai-gateway/clients/claude-code.md`, and its new
relative link still resolves at the new path.
- #30262 now targets `tk/ai-gateway-responses-ws-endpoint` and conflicts
with `main` in 21 files. Against this branch it conflicts in the same 21
files, with `codex.md` at its new path, so this PR doesn't add new
conflicts. Its new relative links still resolve after the move.
- #30805 conflicts in `docs/redirects.json`. It already conflicts with
`main` there, because `main` added the `/docs/about/screenshots` and
`/docs/install/cli` rules.

## Release

No backport to `release/2.39` for now; Nick may apply it by hand later.
A cherry-pick of this branch onto `release/2.39` conflicts in 25 files,
and #31647 and #31650 would need the same treatment to keep the stack
together.
coder/coder.com#1201 and coder/docs-engine#304 merge when the first
release line that carries this layout becomes the docs mainline (2.38
today).

## Known limitations

- **CLI newer than its server.** The `cli/cliui` troubleshooting links
append `/templates/troubleshooting#...` to the server's versioned docs
URL.
A CLI from this release against an older server prints
`/docs/@v2.38.x/templates/troubleshooting#...`.
The router's static
`/docs/templates/:page(open-in-coder|troubleshooting)` rule redirects
that to the old page today, but coder/docs-engine#304 removes that rule
to fix the loop, so after the cutover that case gets the version's
not-found page.

## Post-release follow-up

After a release that includes this stack becomes the default docs
version on coder.com, point these hardcoded URLs at the new paths. They
work today and redirect after the release, so nothing breaks in the
meantime. Anchors are unchanged. Rerun `make gen` afterward to update
`docs/admin/integrations/prometheus.md`,
`scripts/metricsdocgen/generated_metrics`, the CLI golden files, and
`docs/reference/cli/templates/{create,edit}.md`.

| Current (old) URL | Post-release URL | Files |
| --- | --- | --- |
| `https://coder.com/docs/ai-coder/agents` |
`https://coder.com/docs/coder-agents` |
`.claude/skills/doc-check/output.md`, `README.md` |
| `https://coder.com/docs/ai-coder/agents/architecture` |
`https://coder.com/docs/coder-agents/architecture` |
`coderd/x/chatd/chattool/execute_test.go`,
`codersdk/toolsdk/workspacetools/execute.go` |
| `https://coder.com/docs/ai-coder/ai-gateway` |
`https://coder.com/docs/ai-governance/ai-gateway` | `README.md`,
`coderd/templatebuilder/modules/claude-code/module.json`,
`coderd/templatebuilder/modules/codex/module.json` |
| `https://coder.com/docs/ai-coder/mcp-server` |
`https://coder.com/docs/mcp-server` | `server.json` |
| `https://coder.com/docs/admin/templates` |
`https://coder.com/docs/templates` | `README.md` |
|
`https://coder.com/docs/admin/templates/extending-templates/docker-in-workspaces`
|
`https://coder.com/docs/templates/extending-templates/docker-in-workspaces`
| `examples/templates/docker-devcontainer/main.tf` |
|
`https://coder.com/docs/admin/templates/extending-templates/parameters`
| `https://coder.com/docs/templates/extending-templates/parameters` |
`examples/parameters-dynamic-options/README.md`,
`examples/parameters/README.md`, `examples/parameters/main.tf`,
`examples/templates/kubernetes-envbox/README.md`,
`examples/workspace-tags/README.md` |
|
`https://coder.com/docs/admin/templates/extending-templates/prebuilt-workspaces`
|
`https://coder.com/docs/templates/extending-templates/prebuilt-workspaces`
| `coderd/prebuilds/api.go`,
`enterprise/coderd/prebuilds/metricscollector.go` (metric help text) |
|
`https://coder.com/docs/admin/templates/extending-templates/workspace-tags`
| `https://coder.com/docs/templates/extending-templates/workspace-tags`
| `examples/workspace-tags/README.md` |
| `https://coder.com/docs/admin/templates/managing-templates` |
`https://coder.com/docs/templates/managing-templates` |
`cli/templatecreate.go`, `cli/templateedit.go` (CLI help) |
|
`https://coder.com/docs/admin/templates/managing-templates/image-management`
| `https://coder.com/docs/templates/managing-templates/image-management`
| `coderd/templatebuilder/bases/docker/README.md`,
`coderd/templatebuilder/bases/kubernetes/README.md` |
| `https://coder.com/docs/admin/templates/troubleshooting` |
`https://coder.com/docs/templates/troubleshooting` |
`agent/agentscripts/agentscripts.go`,
`coderd/render/escape_internal_test.go`, `codersdk/deployment.go`
(`agentFallbackTroubleshootingURL` default) |
| `https://coder.com/docs/tutorials/template-from-scratch` |
`https://coder.com/docs/templates/template-from-scratch` |
`site/src/pages/TemplateBuilder/TemplateAlternatives.tsx` |

## Validation run

- `make gen/docs-manifest`: clean diff afterward.
- `make lint/docs-manifest`, `lint/docs-redirects` (`go run
./scripts/docsredirectscheck -base origin/main`: 15 rules OK, no
removed-route warning), `lint/docs-html`, `lint/emdash`: pass.
- `pnpm check-docs` (markdownlint + table formatter): pass.
- Vale (`vale --no-exit docs/`): the same 12 warnings and 1 error as
`main`, all pre-existing. Suggestions drop from 646 to 645; the one
removed was in the deleted `ide-agents.md`.
- Local relative-link check over `docs/**/*.md`: 0 unresolved links or
images (also 0 on `main`).
- Pre-commit hook: `make pre-commit-light` on the docs commit, and the
full `make pre-commit` (gen, fmt, lint/go, lint/ts, build, and others)
on the chore commit. Both passed.
- Targeted tests:
- Go: `go test ./coderd/x/chatd/chattool/... ./codersdk/toolsdk/...
./coderd/templatebuilder/...`.
- Site: Vitest for the touched directories (107 files, 1879 tests), and
the story tests for `CoderAgentsProductCard`, `GroupSettingsPageView`,
`GatewayKeysPageView`, and `ModelForm` (4 files, 63 tests).
- Every `docs()` path and anchor in the `site/` and Go diff resolves to
a page and heading in this branch.
- Frontend deep review (all 8 roles) on the `site/` diff of PRs A and B:
no blocking or should-fix findings. Nits on existing code that this PR
only touched, not changed here: a single-use `docsLink` constant in
`ChatAccessDeniedAlert.tsx`, and assertion-only `play` functions in the
`GatewayKeysPageView`, `ModelForm`, `CoderAgentsProductCard`, and
`GroupSettingsPageView` stories.
- Page counts: `docs/**/*.md` (excluding `.style`) 490 → 487;
`manifest.json` routes 495 → 492. With #31646 squash-merged in, the
counts don't change, because the Templates move only moves files.
- **Link health (commit 4):**
- linkspector 0.5.6 with `.github/.linkspector.yml` over all of `docs/`,
like the weekly docs job (`umbrelladocs/action-linkspector`, `reporter:
local`, `filter_mode: nofilter`). I ran it locally with Chrome in
`--no-sandbox` mode.
- `origin/main`: 0 errors. Before the fix (`17e54ad555`): 6 errors, all
`https://coder.com/docs/<new path>` 404s (`shared-workspaces.md:19`,
`esr-2.24-2.29-upgrade.md:80-82`, `tasks-to-chats-migration.md:73`,
`prometheus.md:298`). After the fix (`def074caab`): 0 errors. A first
run after the fix also flagged a Red Hat URL in
`install/server/openshift.md`. That URL is unchanged from `main`, and it
passed on a rerun, so the failure was transient.
- After the fix: `make gen/docs-manifest` (clean), `lint/docs-manifest`,
`lint/docs-redirects` (15 rules OK against `origin/main`),
`lint/docs-html`, `lint/emdash`, `pnpm check-docs`, and the
relative-link check (0 unresolved in 3416 links) pass. The full `make
pre-commit`, including `make gen`, passed on the commit.
  - Tests after the fix:
- Go: `go test ./cli -run 'TestCommandHelp|TestServerYAMLConfig'`,
`./enterprise/cli -run TestEnterpriseCommandHelp`, `./coderd/render`,
`./agent/agentscripts`, `./coderd/prebuilds`,
`./enterprise/coderd/prebuilds`, `./coderd/x/chatd/chattool/...`,
`./codersdk/toolsdk/...`, `./coderd/templatebuilder/...`.
    - Site: Vitest for `src/pages/TemplateBuilder` (7 files, 67 tests).
- **After the rebase onto `ccf5ec74ae`** (commits 5 and 6 included):
`make gen/docs-manifest` (clean), `lint/docs-manifest`,
`lint/docs-redirects` (15 rules OK), `lint/docs-html`, `lint/emdash`,
`pnpm check-docs`, and a relative-link and anchor check over the changed
pages pass.
The only unresolved anchor,
`docs/ai-governance/ai-gateway/cost-controls.md` →
`enterprise.md#list-organization-ai-spend-by-user`, is also broken on
`main`.
The pre-commit hook (`make pre-commit-light`, because both new commits
touch only `docs/`) passed on commits 5 and 6. The rebased code commits
rely on CI for Go and TypeScript checks.
- **Redirect chains:** a Node simulation of coder.com's static rules,
its generated rules from this branch's `docs/redirects.json`, the
router's static rules, the page, then this branch's rules on a miss,
over every page and exact static source, reports the 3 loops above and
nothing else new; with coder/coder.com#1201 and coder/docs-engine#304
applied, 0.
- **Skipped:** the docs-engine sync check, which was not available in
this environment.

<details>
<summary>Implementation notes</summary>

- Rebuilt on `main` at `ccf5ec74ae` (2026-11-03) to clear conflicts with
`main`, after an earlier rebase the same day onto `1a963c628d`. The
previous rebuild was on `9839406ff7` (2026-11-02), and the first version
on `9b8d880d63` (2026-10-23).
- Rebase conflicts with #31603, which added `target="_blank"
rel="noreferrer"` to docs links: 4 `site/` files in the chore commit and
5 in the Templates commit. Each kept `main`'s version of the link with
only the docs path changed. Vitest for the 7 affected page directories
(233 tests) and the story tests for the 8 changed or affected stories
(102 tests) pass, and every `docs()` path in `site/src` resolves except
the same 13 that are unresolved on `main`.
- Rebase conflict: `docs/manifest/coder-agents.yml`. `main` added
`docs/ai-coder/agents/bring-your-own-mcp-servers.md`; it moves to
`docs/coder-agents/` with its sidebar entry, and its
`../../user-guides/` links become `../user-guides/`.
- Folded into the Templates commit: a link that `main` added to
`docs/user-guides/workspace-dotfiles.md` (#31606) now points at
`../templates/extending-templates/environment-variables.md`.
- `git range-diff` against `def074caab`, the head before both rebases,
shows only those changes.
- The move script reads every page from the clean `main` checkout and
resolves each relative link (inline, reference-style, and HTML
`src`/`href`) against the page's **old** directory. It maps the target
through the move map and recomputes the link from the page's **new**
directory. It runs once and refuses to run on a dirty `docs/` tree, so
it can't apply the path math twice. It also rewrites absolute
`coder.com/docs/...` and
`github.com/coder/coder/(blob|edit|tree)/main/docs/...` URLs, and skips
fenced code blocks and inline code.
- Before reusing the script, I ran it on the old base and diffed the
result against the old commit. The only differences were the two
hand-edited glossary entries and table padding from the formatter, so
the rebuild applies the same changes to the current `main`.
- Afterward, `grep -r ai-coder docs/` finds matches only in
`docs/redirects.json`.
- Changes from the earlier version of this branch: 2 more pages move
(`agents/projects.md`, `agents/acp-subagents.md`) and their sidebar
entries move with them. The chore commit also updates
`SessionSummaryTable.tsx` (new on `main`) and the doc-check skill's
`output.md` (its examples moved there from `SKILL.md`). The
`AICostControl.tsx` anchor question is gone, because `main` fixed it.

</details>

---------

Co-authored-by: coder[bot] <137810501+coder[bot]@users.noreply.github.com>
Co-authored-by: Nick Vigilante <nick.vigilante@coder.com>
2026-11-04 17:15:42 -05:00
Mathias Fredriksson c0f7663e95 feat: support editing queued chat messages (#29215)
The owner of a chat can edit a queued message through
`PATCH /api/v2/chats/{chat}/queue/{queuedMessage}`. chatd never sends
a row under edit (`editing_since`): rows ahead of it still send in
order, and a turn that ends at it puts the chat in the new `paused`
status. Ending the head's edit, deleting it, or promoting a queued row
continues the chat. Interrupting sends still run ahead of the queue.

The chat state machine classifies a queue whose head is under edit as
its own sub-state (R1P, I1P, A1P, E1P, XE1P, P), so every path to
`paused` is in the transition matrix. A chat whose family has a paused
member cannot be archived, and auto-archive skips that family. Queue
edit and delete are owner-only, because both can send the next message
with the owner's credentials.

`ChatQueuedMessage` gains `editing_since` and `reasoning_effort`, and
`ChatStatus` gains `paused`.

Closes CODAGT-1028
Refs CODAGT-971
2026-10-30 15:52:22 +02:00
Ethan ea79a9885a fix: return a partial page from read_file instead of an error (#31470)
Closes CODAGT-435

When a `read_file` response would go over 32 KB, the agent returned an
error and no content, telling the model to "Read less at a time using
offset and limit parameters". That's most source files over about 32 KB,
so the model would lose a turn on each one just to pick a smaller range.
#22400 made these hard errors on purpose, so the model wouldn't answer
from partial data, but most of the prior art I looked at (pi-mono,
Gemini CLI, opencode, Cline) just returns what fits and says where to
continue. I think that handles the partial data concern better, as the
model is told exactly which lines it has.

So, the agent now ends the page at whichever comes first of `limit`,
2000 lines, 32 KB, or the end of the file, and returns it as a success.
`workspacetools.ReadFile` then adds a `note` to any page that stops
before the end of the file, so the chat tool and the MCP tool both get
it. Reading `agent/agentfiles/files.go` with no `offset` or `limit` used
to error, and now gives 956 lines with:

> Showing lines 1-956 of 1576. To read more, call this tool again with
offset=957.

The issue suggested returning a chunk plan (e.g.
`[[0,200],[200,600],...]`) instead. The model needs one call per page
either way, and a chunk list would depend on the current limits and line
lengths, whereas the next offset is just `offset + lines_read`.

A `limit` over 2000 is now clamped rather than erroring, and
`total_lines` no longer counts a trailing newline as an extra line,
since the note's maths depends on it. I also dropped the "use offset and
limit to read a portion" hint from the 1 MB error, as that check runs
before `offset` and `limit` are applied. Workspaces running an older
agent will keep returning the error until their agent is updated.

I tested this on a dev deployment through Coder Agents, asking Claude
Haiku 4.5 and GPT-4.1 mini to read a 108 KB and a 118 KB file (3000 and
3200 lines) using only `read_file`, then report the last line. Both
followed the note through to the end of the file (Haiku in 4 calls,
GPT-4.1 mini in 5) and got the last line right, whereas on `main` the
first call just errors. `coder_workspace_read_file` over MCP gives the
same note. The only odd behaviour I saw was Haiku once skipping straight
to the end of a file when the prompt only asked about the last line,
which I think is on the model rather than the tool.
2026-10-30 12:23:17 +11:00
Michael Suchacz c7f1f4b9ef refactor: delete the ChatProviderAPIKeys test seam (#30927)
This PR deletes `coderd.Options.ChatProviderAPIKeys`, its
`coderdtest.Options` copy, and the
`coderdtest.OpenAICompatProviderAPIKeys` and
`coderdtest.FakeOpenAICompatProviderAPIKeys` helpers. Only tests set the
option. `newChatTestOptions` and 27 other chat, MCP, and toolsdk
fixtures used it to start a fake OpenAI-compatible server whose fallback
keys no model call reads.

## Why it is safe

- Model calls never read the fallback keys. `newModel` builds the client
from `fantasyConfigForAIBridge`, which returns AI Gateway placeholder
keys (`coderd/x/chatd/model_routing_aibridge.go`).
- Key resolution for a selected provider passes empty fallback keys
(`resolveUserProviderAPIKeysForProvider` in `chatd.go`), and so does the
HTTP availability check in `coderd/exp_chats.go`. Active model configs
always have an `ai_provider_id` (the
`chat_model_configs_ai_provider_required_when_active` check constraint).
- The only path that merges `chatd.Config.ProviderAPIKeys` is the
provider-type lookup for the computer-use subagent. It asks only for
`anthropic` or `openai`, and the test keys were stored only under
`openai-compat`.
- Model-backed tests get their key and base URL from the DB provider
that `CreateOpenAICompatChatModel` creates.
- `git grep -n 'ChatProviderAPIKeys\|OpenAICompatProviderAPIKeys'`
matches nothing after the change.

## Review focus

`coderd.go` no longer passes `ProviderAPIKeys` to `chatd.Config`.
Production never set `ChatProviderAPIKeys`, so chatd gets the zero value
it always got in production. `chatd.Config.ProviderAPIKeys` stays for
now, because chatd internal tests still set it directly.

## Validation

Each suite ran before the change (on `e289c6fe011`) and after it, with
the same results:

- `go test ./coderd/ -run '(?i)chat|mcp' -count=1 -p 4 -parallel 4 -v`:
pass, 163 top-level tests, 899 PASS lines, 0 failures, same test set
both times.
- `go test ./codersdk/toolsdk/ -count=1 -parallel 4 -v`: pass, 234 PASS
lines, 0 failures.
- `go test ./coderd/coderdtest/... -count=1 -v`: pass, 350 PASS lines, 0
failures.
- `go build ./...`, `go vet` on the touched packages and
`./enterprise/coderd/...`, and `golangci-lint run --fix=false ./coderd/
./coderd/coderdtest/... ./codersdk/toolsdk/...`: clean.
- `make pre-commit`: passed.

> Xum, an AI agent, opened this PR for @ibetitsmike and wrote this
description.
2026-10-27 10:15:44 +01:00
coder[bot] 69fab7de06 chore!: remove the coder_report_task MCP tool (#30742)
Coder Tasks have been removed from coder/coder, so the
`coder_report_task` MCP tool and the status reporting pipeline behind it
no longer do anything useful. This PR removes the tool and the code that
only existed to support it.

### Removed

- `codersdk/toolsdk`: `ToolNameReportTask`, `ReportTaskArgs`,
`ReportTask`, `Deps.report`, `WithTaskReporter`, and the `All` entry,
plus their tests.
- `coderd/mcp`: the remote MCP exclusion for `coder_report_task`.
- `coder exp mcp server`: the task reporter, the AgentAPI screen
watcher, the agent socket connection, and the `coder_report_task` skip
logic. The server now requires an authenticated user client, because
every remaining tool needs one.
- `coder exp mcp configure claude-code`: the
`mcp__coder__coder_report_task` allowed tool, the injected
`coder_report_task` CLAUDE.md prompt, the `report tasks properly`
history line, the default system prompt asking the agent to send task
status updates, and the `CODER_MCP_APP_STATUS_SLUG` /
`CODER_MCP_AI_AGENTAPI_URL` env written to the Claude MCP config.
- `cli/cliutil/queue.go`: only used by the reporter.
- `github.com/coder/agentapi-sdk-go` (and transitive `tmaxmax/go-sse`)
via `go mod tidy`.
- The "task reporting" mention in `docs/ai-coder/mcp-server.md`.

### Breaking: removed `coder exp` flags and env

`coder exp` is experimental. Any scripts that still pass these flags
will now fail with an unknown flag error. Env vars that are no longer
read are ignored.

| Command | Flag | Env |
|---|---|---|
| `coder exp mcp server` | `--app-status-slug` |
`CODER_MCP_APP_STATUS_SLUG` |
| `coder exp mcp server` | `--ai-agentapi-url` |
`CODER_MCP_AI_AGENTAPI_URL` |
| `coder exp mcp server` | `--socket-path` | `CODER_AGENT_SOCKET_PATH`
(still used by `coder agent` and `coder exp sync`) |
| `coder exp mcp configure claude-code` | `--claude-app-status-slug` |
`CODER_MCP_APP_STATUS_SLUG` |
| `coder exp mcp configure claude-code` | `--ai-agentapi-url` |
`CODER_MCP_AI_AGENTAPI_URL` |

`--claude-system-prompt` no longer has a default.

<details>
<summary>Intentionally kept</summary>

- **Workspace app status reporting** (`UpdateAppStatus` on the agent API
and agent socket, `agentsdk.PatchAppStatus` / `ProtoFromPatchAppStatus`,
the deprecated `PATCH /workspaceagents/me/app-status` endpoint): still
served by coderd and the agent and part of versioned protocols. With
this change, nothing in the CLI calls the agent socket `UpdateAppStatus`
client anymore. Removing it would need a separate protocol change.
- **`report_tasks` in
`coderd/templatebuilder/modules/{aider,amazon-q}`**: these catalog
entries mirror the input variables of pinned coder/registry module
versions, and the behavior lives in those modules. Dropping the variable
here would not disable it, because `amazon-q` defaults it to `true`. The
registry modules need a follow-up instead.
- **`UserClientOptional` on toolsdk tools**: other tools still set it,
and remote MCP still uses it.
- No CLI golden files or generated CLI docs changed, because `coder exp`
commands are hidden.

</details>

### Verification

- `go build ./...`
- `go test ./codersdk/toolsdk/... ./coderd/mcp/... ./cli/cliutil/...`
- `go test ./cli -run 'TestExpMcp|TestCommandHelp'`
- `golangci-lint run ./cli/... ./codersdk/toolsdk/... ./coderd/mcp/...`
(0 issues)

---

_Generated by Coder Agents._

Co-authored-by: coder[bot] <137810501+coder[bot]@users.noreply.github.com>
2026-10-14 16:34:56 +05:00
Atif Ali 261269aff0 fix: support path-aware OAuth discovery for remote MCP (#30168)
## Problem

VS Code discovers Coder's MCP endpoint as a path-qualified OAuth
resource, but Coder advertises the deployment root. Accepting MCP-bound
tokens also requires keeping REST-backed tool calls within the
originating MCP resource rather than granting those tokens direct REST
access.

## Fix

- Publish path-aware protected resource metadata and authentication
challenges; keep authorization-server metadata at the root.
- Validate the MCP audience and delegate tool requests through a private
in-process transport that retains identity, scopes, expiry, revocation,
and streaming cancellation.
- Reject unsupported MCP path aliases and validate SDK path components
before URL resolution.
- Add discovery, audience, authorization, endpoint-isolation, and
transport lifecycle regression tests and update integration docs.

## Validation

VS Code completed OAuth and returned the authenticated user through a
real MCP tool. Official MCP SDK 1.30.1 completed discovery,
registration, PKCE, tool execution, and refresh. These live checks
preceded the final path hardening; the hardened paths are covered by
automated regressions.

Affected Go package tests, targeted race tests, backend/SDK lint, and
targeted documentation checks passed during self-review. The exact PR
commit was also checked separately from the uncommitted consent-scope
fix.

> [!NOTE]
> Tokens are MCP-audience-bound, but grants still include `coder:all`;
least-privilege scopes remain a separate requirement
([PLAT-639](https://linear.app/codercom/issue/PLAT-639)).

Preview:
https://coder.com/docs/@fix/mcp-internal-delegation/ai-coder/mcp-server

<details>
<summary>Implementation decisions and review context</summary>

1. Treat `/api/experimental/mcp/http` and its single-trailing-slash form
as supported MCP resource paths; reject other aliases and descendants
before route authentication.
2. Preserve deployment-root discovery and authorization-server metadata
for existing REST integrations.
3. Bind private delegation to the authenticated API-key ID. Revalidate
the original credential and enforce normal RBAC/scopes on every internal
REST request; do not derive delegation from HTTP headers.
4. Keep internal HTTP, streaming, and WebSocket traffic in-process with
no network fallback. Close the transport on outer-request cancellation
and normal completion.
5. Validate workspace owner/name path segments before SDK URL
construction, including tool-level tests that malformed arguments cannot
select a different REST endpoint.
6. Keep compatibility and delegated tool execution together:
discovery-only changes would allow authentication while leaving real MCP
tool calls broken.

</details>

> 🤖 This PR was created with the help of Coder Agents, and needs a human
review. 🧑💻
2026-10-09 00:58:39 +05:00
Mathias Fredriksson cc76b62009 refactor!: drop edit_files search/replace wire compat (#29221)
Remove the deprecated `search`/`replace` keys from the `edit_files`
wire type. `FileEdit` now encodes and decodes `old_text`/`new_text`
only; the shims that carried the old keys through the rename (#28467)
are gone.

**Breaking change:** chat lifecycle hook consumers must use `old_text`
and `new_text` for `edit_files`. A `pre_tool_use` `input_override`
that still sends `search`/`replace` fails with `old_text must not be
empty`.

Fixes CODAGT-483
2026-10-06 10:42:44 +00:00
Kyle Carberry 98332c739b feat: align MCP workspace tools with Coder Agents and add execute tools (#30275)
The control-plane MCP server's workspace tools now share their
arguments, behavior, and result formats with the Coder Agents tools. The
MCP server also exposes the Coder Agents execute and process tools.

The shared logic lives in a new `codersdk/toolsdk/workspacetools`
package that both `coderd/x/chatd/chattool` and `codersdk/toolsdk` call.
The Coder Agents tool schemas, descriptions, and JSON responses are
unchanged and the existing chattool tests pass unmodified. The one
exception is the missing-shell guidance, which now reads "Workspace
commands run with `sh -c`" instead of naming Coder Agents. Golden files
in `coderd/mcp/testdata/workspace_tools` hold the shared contract for
each tool, and a test checks both the Coder Agents and MCP tools against
them, so contract changes show up in review. The test also fails if MCP
text names a tool that only exists in Coder Agents.

The MCP versions differ from Coder Agents in four ways:
- Each tool takes a `workspace` argument.
- Invalid arguments and agent API errors come back as MCP tool errors.
- A single call waits at most 5 minutes, matching the agent's maximum
blocking wait. Longer timeouts are rejected with a pointer to
`run_in_background`.
- There is no `model_intent` argument.

**Breaking changes to `/api/experimental/mcp/http` and `coder exp
mcp`:**

| Before | After |
|---|---|
| `coder_workspace_read_file`: byte offset/limit, base64 `content` +
`mimeType`, binary-safe | Line-based `offset`/`limit`, text only,
returns `{content, file_size, lines_read, total_lines}` |
| `coder_workspace_write_file`: base64 `content`, binary-safe |
Plain-text `content`, returns `{ok: true}` |
| `coder_workspace_edit_file` + `coder_workspace_edit_files` |
`coder_workspace_edit_files` only, returns `{files: [{path, diff}], ok}`
|
| `coder_workspace_bash` over SSH: `timeout_ms` (default 60s, 5m cap),
kills the command on timeout (exit 124), trims output |
`coder_workspace_execute`: `timeout` (default 10s, 5m cap), `workdir`,
`run_in_background`. On timeout the command keeps running and the result
has `background_process_id`. Output is untrimmed and capped at 32KB. |
| None | `coder_workspace_process_output`,
`coder_workspace_process_list`, `coder_workspace_process_signal` |

`--allowed-tools` lists and templates that name the removed tools will
need updating. A code search found `coder_workspace_bash` in a couple of
demo templates (`coder/demo-env-templates`, partner demos).

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

- **Shared core vs. duplication.** The execute lifecycle has the most
logic that could drift: waiting in the foreground, handing off to the
background on timeout, recovering a snapshot after transport errors, and
truncating output. It's extracted rather than copied, along with
edit-file validation, read-file handling, and result types. Each surface
keeps its own protocol handling: fantasy and MCP response envelopes,
plan-turn restrictions, `model_intent`, and workspace resolution and
authorization. Struct-tag descriptions must be literals, so the
golden-file test is what keeps them in sync with the shared constants.
Shared result structs keep their fields in alphabetical order so the
JSON matches what Coder Agents produced from maps before.
- **Replace instead of add alongside.** Keeping `coder_workspace_bash`
next to `coder_workspace_execute` would give models two overlapping ways
to run commands. The read and write contracts change anyway, so keeping
the old names wouldn't preserve compatibility. The endpoint and CLI
command are both experimental.
- **5-minute cap on MCP calls.** Each stateless MCP request has to
return before the client gives up. Otherwise the `background_process_id`
needed to follow up is lost and the command keeps running with no ID the
model knows about. Coder Agents has no cap, so its behavior is
unchanged.
- **No SSH fallback.** `execute` doesn't fall back to SSH on older
agents, because SSH can't support the process management contract. The
agent process API and line-based reads shipped in February 2026.
- **Process scoping.** The agent scopes processes by chat ID, which
coderd sets only for Coder Agents connections. MCP calls carry no chat
ID, so they can't see or signal Coder Agents chat processes. All MCP
callers share one process namespace per workspace, which means anyone
with SSH access can list and signal processes another MCP user started.
That matches what SSH access already allows. The existing `ActionSSH`
check in `coderd/mcp_http.go` still runs before every agent connection.
Nothing reaps MCP processes; they run until they exit, are signaled, or
the agent stops.
- **Errors.** As in Coder Agents, command failures and timeouts come
back as result JSON (`success: false`, `background_process_id`). Agent
API errors have transport metadata removed, the same way Coder Agents
formats them.

</details>

---

🤖 Generated by Coder Agents on behalf of @kylecarbs.
2026-10-04 11:36:48 -05:00
Mathias Fredriksson ff158c0589 feat: preserve user-supplied chat titles (#29486)
Chats created through the API or MCP couldn't be given a title, and
automatic title generation could overwrite a title the user had set,
at creation or by renaming during generation. #30254 tried to keep
renames by comparing title text, but text can't tell a user title
from a generated one that has the same text.

Chat creation over the API and MCP now accepts a title. Each chat
records where its title came from: fallback (derived from the
prompt), generated, or user. A title can only be replaced by a title
from the same source or a higher one, so automatic generation never
overwrites a user's title. Title changes no longer reorder the chat
list, and clients apply them in the order they were written. Chats
created before this change are treated as having fallback titles.

Closes CODAGT-1047
Reverts #30254
2026-10-02 12:58:53 +00:00
Jake Howell d853893e8a feat: filter agent chats by chat status (#30009)
> 🤖 This PR was written by Coder Agents on behalf of Jake Howell.

The sidebar's `Chat status` filter only offered `Read`/`Unread`. This
turns it into a real status filter backed by the `chat_status` enum, the
same value the row icon already shows. Stacked on #30006.

- New `status:<waiting|running|error|requires_action|interrupting>`
search filter (repeated or comma-separated) plus a `chat_statuses` param
on `GetChats`.
- Sidebar `Status` submenu offers `Requires action`, `Error`, `Working`
and `Idle`; `Working` covers `interrupting`, since stopping a turn is
too brief to be worth its own option. A full selection emits no filter.
- `Unread` moves out to its own checkbox (`unread=true` in the URL), so
it narrows a status selection instead of competing with it.
- Grouping by status now sections by the five statuses instead of
`Unread`/`Read`, reusing `getChatStatusDisplay` for labels and icons.

<img width="439" height="177" alt="image"
src="https://github.com/user-attachments/assets/82ef2ef0-cde9-4169-bb29-caa7b9bfdb9e"
/>
2026-09-28 19:58:35 +10:00
Jake Howell 4f2fd36e8f feat: add pr_status:none filter for chats without a pull request (#30006)
> 🤖 This PR was written by Coder Agents on behalf of Jake Howell.

Adds `none` to the `pr_status` chat filter, matching chats with no
linked pull request. Stacked on #30005.

- `none` means no diff-status row, or a row whose `pull_request_state`
is null/empty (so chats whose PR was cleared still match), and it ORs
with the other statuses, e.g. `pr_status:none,open`.
- Frontend picks it up from the shared `CHAT_LIST_PR_STATUS_ORDER`,
shown as `No PR`; `searchQuery.ts` now derives its valid set from that
list instead of a duplicated literal.
- Swagger, API docs and the `list_chats` tool schema
regenerated/updated.

<img width="508" height="464" alt="image"
src="https://github.com/user-attachments/assets/bd6c70e3-c634-499d-ae16-e7953f6da2f9"
/>
2026-09-28 19:14:12 +10:00
Michael Suchacz b0b472de9e feat(codersdk/toolsdk): add owner_id to the coder_create_chat MCP tool (#29704) 2026-09-25 09:43:07 +02:00
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
Michael Suchacz 9cd38b5373 feat!: remove the Tasks API from coderd and codersdk (#28789)
Removes the Tasks API and task orchestration from coderd and codersdk:
the `/api/v2/tasks` handlers, task middleware and search filters, task
name generation, autobuild pause/resume logic, task notification
enqueues, telemetry, and the codersdk task types. Regenerates apidoc,
the API reference docs, and `typesGenerated.ts`.

**Breaking change:** the Tasks API is removed; requests to
`/api/v2/tasks` fail. Tasks was deprecated in v2.34.

**Stack context:** Coder Tasks removal stack (CODAGT-642): #28743
frontend → #28788 CLI → #28789 API → #28790 provisioner → #29084
notifications → #29085 build reasons → #29086 tables → #28750
permissions. The last four PRs replace the single data-drop migration
that used to live in #28750; each carries one concern and its own
migration number.

**Why:** with the UI (#28743) and CLI (#28788) gone, this deletes the
server surface. Database tables, RBAC, and notification templates stay
until the top of the stack so every layer builds and migrates
independently.

**What to scrutinize:**
- The `enable-ai-tasks` option is removed outright (review feedback):
configs still carrying `enableAITasks` fail to parse until the key is
deleted.
- provisionerdserver now writes `HasAITask: false` on template/build
flags; the column and these lines are dropped in #29085.
- `codersdk.ResourceTypeTask` stays so audit logs recorded before the
removal keep rendering.

**Validation:** `go build ./...`, `go vet ./...`, `go test
./codersdk/...`, site biome and tsc, CLI goldens regenerated, pre-commit
hooks.

**Size:** +242/-10,792 over 76 files; additions are regenerated
docs/goldens and test fixture updates.

> Created by Xum (AI agent) on behalf of @ibetitsmike.
2026-09-11 01:48:56 +00:00
Hugo Dutka 8e3c36ec5a fix: standardize org use in the mcp server tools (#29141)
Addresses
https://linear.app/codercom/issue/CODAGT-970/expose-mcp-tools-for-organization-selection.

Currently our MCP server does not handle organizations consistently.
Some tools, like `create_template`, use a random organization if the
user is in multiple orgs. Others, like `create_chat`, pick the
organization which contains the latest chat the user spawned.

This PR standardizes organization use in the MCP server tools.
Organization ID is now in the argument schema of all tools that depend
on it. It's always optional. If the user is only a member of one
organization, tools default to that organization. If the user is a
member of multiple organizations, tools return an error asking to supply
an organization id.

This PR also adds a new `list_organizations` tool to allow MCP clients
to view the organizations that the user has membership in.
2026-09-10 11:48:08 +02:00
Hugo Dutka 19b72a7bed fix(codersdk/toolsdk): remove background test startup timing assumptions (#29043)
The background bash tests assumed their initial output would arrive
within a two-second deadline. Separate completed-command output capture
from timeout behavior, and gate background completion until after the
handler returns. Each run now uses its own temporary FIFO and completion
marker instead of a fixed sleep and shared file.

Fixes https://linear.app/codercom/issue/CODAGT-1004

<details>
<summary>Investigation and validation evidence</summary>

Injecting a three-second delay into external command execution
reproduced both original failures with exit code 124 and no initial
output. This demonstrates the unsafe startup-timing assumption; it does
not establish the precise source of the CI runner's delay or implicate
the output-drain window.

The updated tests passed three repetitions with the same injected delay,
five race-enabled repetitions, the full `toolsdk` package, and package
lint checks (zero issues).

The continuation test releases its FIFO only after the timeout result
and verifies the unique completion marker afterward. Output capture is
covered separately for stdout and stderr.

</details>

Generated by Coder Agents on behalf of @hugodutka.
2026-09-08 16:15:05 +02:00
Hugo Dutka fcaa14f456 fix: validate mcp input (#28702)
Addresses
[CODAGT-994](https://linear.app/codercom/issue/CODAGT-994/return-validation-errors-for-invalid-mcp-tool-arguments).

The MCP server currently does not validate tool call arguments against
the MCP JSON schema.

Additionally, when a client passes invalid arguments to a Coder MCP
server tool call, currently the server returns only:

```json
{
  "error": "MCP tool request failed."
}
```

In this PR we start validating arguments and returning an informative
validation error instead:

```json
{
  "error": "invalid arguments: ...",
  "expectedSchema": { ... }
}
```
2026-09-07 19:15:38 +02:00
Michael Suchacz 8acd6196b7 fix(codersdk/toolsdk): align chat model listing org default with chat creation (#28762) 2026-08-31 06:27:00 +02:00
Mathias Fredriksson f208fc0c83 refactor: rename edit_files wire type to old_text/new_text (#28467)
The schema has advertised old_text/new_text since #25658, but the
shared wire type still encoded search/replace.

Decode accepts the deprecated keys only when both new fields are
empty, through the pre-rename tags, so old coderd and cached MCP
schemas keep working. Marshal emits both key sets so old agents keep
working after a coderd upgrade. Both shims are removed in the first
release after Coder Agents GA (2026-09).

The Go field rename is a deliberate source break: FileEdit carries
no codersdk stability guarantee.

Fixes CODAGT-523
Refs CODAGT-483
2026-08-27 16:06:21 +03:00
McKayla はな 2a8e390f0c refactor: replace most usage of ptr.Ref with new (#28282) 2026-08-26 11:57:56 -06:00
Michael Suchacz 973d5a458b feat: promote codersdk chat API methods to Client (#28497)
## Stack context

This is the second PR in the 3-PR chat API promotion stack: server
compatibility mounts (#28496), codersdk promotion (this PR), and
frontend path updates (#28498).

## Summary

Move the promoted chat and MCP SDK methods from `ExperimentalClient` to
`Client` and update them to use `/api/v2`. Methods for routes that
remain experimental stay on `ExperimentalClient`.

Update in-repository callers and generated types. The multi-replica chat
stream relay dials `/api/v2` directly: mixed-version replica sets are
not a supported upgrade path, so no experimental-path fallback is kept
(per review). Remote dogfood UAT passed for the composed stack.

> [!NOTE]
> Xum acted on Mike's behalf in this pull request.
<!-- xum-attribution: model=claude-fable-5 thinking=high -->
2026-08-26 18:46:34 +02:00
Hugo Dutka 31146642ba fix(codersdk/toolsdk): select default chat org based on recent usage (#28543)
When `coder_create_chat` omitted `organization_id`, it selected the
first organization membership, which could choose an organization
without an available chat model.

Default to the organization used by the user's most recently updated
chat. For users with multiple organizations and no prior chats, return a
descriptive error requiring `organization_id`; single-organization
behavior is unchanged.

Closes
https://linear.app/codercom/issue/CODAGT-968/fix-mcp-chat-creation-organization-selection

<details>
<summary>Coder Agents disclosure</summary>

This pull request was generated by Coder Agents.
</details>
2026-08-26 14:52:44 +02:00
Ethan aefc761fcf feat!: enforce organization scope for chat models (#27959)
## Summary

Move chat model management and runtime availability to
organization-scoped API routes. The API now identifies the organization
in every model-management operation, applies chat model RBAC and ACLs
within that organization, and returns only models the caller can use
there.

Depends on #27958.

## Problem

Chat model configurations now belong to organizations, but the existing
experimental management and discovery APIs are deployment-shaped. Their
routes do not identify an organization, SDK methods do not accept an
organization ID, and item operations address a model configuration by ID
alone.

That contract cannot safely expose organization-owned model
configurations. It also does not provide the organization-specific
provider information, authorization behavior, ACL management, or
recovery guidance needed by the settings and chat clients later in the
stack.

## Fix

Add organization-scoped chat model endpoints under:

```text
/api/experimental/organizations/{organization}/chats/models
```

The new API provides:

- Organization-scoped model listing and creation.
- Model get, update, and delete operations scoped by both organization
and model ID.
- Model ACL read and update operations.
- Organization-scoped runtime availability and provider status.
- Redacted provider descriptors for model-management clients.
- Chat model RBAC and row ACL enforcement for writes, runtime discovery,
and Chatd model selection.
- Recovery guidance when a chat's organization has no usable local
model.

Historical chats can retain a model reference owned by another
organization, but runtime selection uses an authorized local default
when one is available. Foreign, inaccessible, disabled, and otherwise
unusable models are not exposed as usable selections for the
organization.

The existing collection and availability routes remain as
default-organization compatibility endpoints in this layer:

```text
GET  /api/experimental/chats/model-configs
POST /api/experimental/chats/model-configs
GET  /api/experimental/chats/models
```

These compatibility routes do not provide access to models in
non-default organizations. The previous unscoped item update and delete
routes are removed because item operations must include the owning
organization.

## Breaking change

This changes the experimental HTTP API and Go SDK contracts for chat
model management.

Clients must migrate model-management and availability requests to the
organization-scoped routes and provide an organization identifier. In
the Go SDK, the following methods now require an organization ID:

- `ChatModelAvailability`
- `ChatModels`
- `CreateChatModel`
- `UpdateChatModel`
- `DeleteChatModel`

`ChatModels` also changes its return value from `[]ChatModel` to
`OrganizationChatModelsResponse`, which includes the organization models
and the redacted provider descriptors required by management clients.
New item and ACL methods likewise require both the organization ID and
model ID.

The default-organization compatibility routes reduce the immediate
impact for collection and availability callers, but they do not preserve
the previous item API or SDK method signatures. Generated clients and
direct consumers of the experimental API must be updated before adopting
this change.

_This pull request description was generated by Coder Agents._
2026-08-24 11:20:25 +01:00
Ethan 9bd8899359 refactor: clarify chat model resource names (#27958)
Rename the admin-managed SDK resource from `ChatModelConfig` to
`ChatModel`. Rename catalog entries from `ChatModel` to
`ChatModelCatalogEntry` and replace `ChatModelsResponse` with
`ChatModelAvailabilityResponse`.

API routes and JSON fields do not change. Experimental SDK and OpenAPI
schema names do change.

Depends on #27957

_This pull request description was generated by Coder Agents._
2026-08-24 09:43:45 +00:00
McKayla はな 5ca0143992 refactor: upgrade golangci-lint to v2.0.2 (#28416) 2026-08-21 23:46:27 -06:00
Michael Suchacz 2f996d765a feat: remove task management MCP tools (#28354)
Removes the six task-management tools from the MCP tool surface in
`codersdk/toolsdk`:

- `coder_create_task`
- `coder_delete_task`
- `coder_list_tasks`
- `coder_get_task_status`
- `coder_send_task_input`
- `coder_get_task_logs`

Both the coderd-hosted MCP server (`coderd/mcp`) and the CLI `coder exp
mcp server` derive their tool list from `toolsdk.All`, so removing the
registry entries removes the tools from both servers with no server-side
changes.

The MCP server documentation (`docs/ai-coder/mcp-server.md`) is updated
to drop the task-management entry from the tool overview.

## What stays

- `coder_report_task` is intentionally kept: it is the agent-side
status-reporting mechanism (`WithTaskReporter`), already excluded from
the hosted MCP server, and required by `coder exp mcp server`, Claude
Code configuration, and scaletest.
- The `codersdk` Task SDK methods (`CreateTask`, `DeleteTask`, `Tasks`,
`TaskLogs`, etc.) and the Tasks feature (CLI, UI, API) are untouched;
this PR only removes MCP tool registration.
- The task-only `taskIDDescription` helper is deleted; `userDescription`
stays (used by `coder_create_workspace`).

The code change is deletion-only: 970 removed lines across `toolsdk.go`
and its test file, no added behavior.

## Validation

- `go build ./...`, `go test ./coderd/mcp/`, and targeted
`golangci-lint` pass. `go test ./codersdk/toolsdk/` passes all tests;
the package-level goleak failure (agentssh goroutines) reproduces
identically on unmodified `main` and is unrelated to this diff.
- `pnpm run lint-docs` passes on the docs change.
- Remote dogfood UAT ran against the toolsdk change and passed:
`tools/list` on both the hosted MCP endpoint and `coder exp mcp server`
confirms the six tools are absent and the remaining surface is intact
(34 hosted tools; CLI adds `search`/`fetch`), a `tools/call` of
`coder_get_authenticated_user` succeeds, and calling a removed tool
returns `-32602 unknown tool`. `coder_report_task` still registers in
the CLI server when `CODER_MCP_APP_STATUS_SLUG` is set.
- The failing `Pixel / Review` commit status is unrelated: this PR
contains no `site/` changes, and the same status fails on the last 8
`main` commits including this PR's merge base (`676b9bb06bf`).

> Mux acted on Mike's behalf to create this pull request.
2026-08-20 12:07:30 +02:00
Michael Suchacz affeeaf9c8 feat: extend agent chat MCP tools for remote UAT evidence loops (#28233)
Extends the Agent-chat MCP tools so an unattended UAT evidence loop can
fetch artifacts, monitor long runs, and find prior runs without burning
model context.

## Backend

- New `chat_files_token` crypto key feature (migration 000571) with
rotator support and a dedicated signing keycache on coderd.
- `POST /api/experimental/chats/files/{file}/download-url`
(authenticated) mints a short-lived (5 min) signed URL and returns it
with `sha256`, `size_bytes`, `name`, `mime_type`, and `expires_at`.
- `GET /api/experimental/chats/files/{file}/download?token=` (no session
token) redeems the signed URL: verifies the JWS, requires the token's
`file_id` to match the path, and re-checks the minting user's RBAC
access live at redemption. Clients can `curl -o` artifacts with zero
credentials in the URL consumer.
- `ChatFileMetadata` gains `size_bytes` (via `octet_length`, no bytes
fetched).

## MCP tools (`codersdk/toolsdk`)

- `coder_download_chat_file`: by `file_id` or `chat_id`+`file_name`;
returns the signed URL plus checksum and size instead of base64.
- `coder_await_chat`: blocks (bounded `wait_secs`, 1-120) until a chat
leaves `running`/`interrupting`, using the existing watch stream with
subscribe-before-read.
- `coder_list_chats`: label, query, and limit filtering; chat
projections now include labels.
- `coder_get_chat_messages`: `after_id` forward cursor with
`next_after_id` (exact incremental reads), plus per-message `files`
metadata so artifact-bearing messages are identifiable.
- `coder_get_chat`: file listings now include `size_bytes` and
`created_at`.
- `coder_list_templates`: exposes `agents_allowed` for pre-flight
checks.

## Testing

- coderd: mint/redeem happy path with an unauthenticated client,
expired/tampered/file-mismatched tokens, auth still required on the
plain file endpoint, non-owner mint rejection.
- toolsdk: harness + integration coverage for all new/changed tools,
including signed-URL redemption with checksum verification,
forward-cursor exactness, await transition/timeout paths, and label
filtering.
- Remote dogfood UAT (dev.coder.com Coder Agent) passed all six
acceptance scenarios end to end over both MCP transports.

Note: `go test ./codersdk/toolsdk/` has a pre-existing goleak flake on
main (leaked `agentssh` non-PTY session goroutines from SSH exec tests;
reproduced 3/3 on clean `b4971bc49f1`). It is unrelated to this diff.

> Mux acted on Mike's behalf to create this PR.

<!-- mux-attribution: model=claude-sonnet-4-6 thinking=high -->
2026-08-18 19:15:30 +02:00
Michael Suchacz 8d4d0b35dd feat: add Coder Agents chat tools to the MCP toolsdk (#28025)
Exposes the experimental Coder Agents chats API through the MCP tool
registry, so MCP clients (the hosted `/api/experimental/mcp/http` server
and `coder exp mcp server`) can start and drive server-side coding
agents.

New tools in `codersdk/toolsdk`, all thin wrappers over existing
`codersdk.ExperimentalClient` methods:

| Tool | Wraps |
|---|---|
| `coder_create_chat` | `CreateChat` (prompt, optional org, model
config, labels) |
| `coder_get_chat` | `GetChat` (status, last error, last turn summary,
workspace, files) |
| `coder_get_chat_messages` | `GetChatMessages` (user-facing parts,
chronological, cursor pagination, queued prompts) |
| `coder_send_chat_message` | `CreateChatMessage` (queue or interrupt
busy behavior) |
| `coder_interrupt_chat` | `InterruptChat` |
| `coder_archive_chat` | `UpdateChat` with `archived: true` |
| `coder_list_chat_model_configs` | `ListChatModelConfigs` (enabled
configs with default flag) |

Both MCP servers register tools from `toolsdk.All`, so no additional
wiring is needed. Responses are trimmed to what an MCP caller needs (IDs
as strings, user-facing transcripts) rather than full SDK payloads. No
new endpoints and no database changes.

Also adds MCP
[prompts](https://modelcontextprotocol.io/specification/2026-07-28/server/prompts)
for the chat workflows, defined once in `codersdk/toolsdk` and
registered by both servers:

| Prompt | Purpose |
|---|---|
| `coder_agents_delegate` | delegate a task to a Coder Agents chat and
monitor it to completion |
| `coder_agents_check` | check the status and recent activity of an
existing chat |

Each prompt declares the tools its workflow needs; the stdio server
skips prompts whose tools are excluded by `--allowed-tools`.

Tests run the tools against a chat-enabled coderdtest instance (fake
OpenAI-compatible provider plus in-process AI bridge), covering the full
lifecycle, an interrupt against a blocked turn, pagination cursors,
permission-dependent model config filtering, and argument validation.
Prompt coverage spans SDK rendering, the hosted
`prompts/list`/`prompts/get` round trip, and the stdio server including
allowlist gating.

> Mux created this PR on Mike's behalf.
2026-08-13 18:32:47 +02:00
Mathias Fredriksson 9fc12afdaa test(codersdk/toolsdk): use portable echo in WorkspaceSSHExec test (#25840)
PowerShell's echo aliases to Write-Output, which rejects -e as
an ambiguous parameter and exits 1. Use plain echo with spaces
instead. Remove the Windows t.Skip and TestMain exception.

TestMain untested-tools check now only fails on full-suite runs.
Filtered runs (e.g. -run TestTools) warn instead.

Closes CODAGT-518
2026-06-01 18:01:19 +03:00
Mathias Fredriksson 1ddc89caa2 test(codersdk/toolsdk): scope err in SendTaskInput and GetTaskLogs subtests (#25434)
Fixes coder/internal#1475
Fixes CODAGT-364
2026-05-20 11:02:46 +03:00
Garrett Delfosse 78d4cf9e47 fix: soft-delete stale workspace agents on new build (#25207) 2026-05-18 08:33:29 -04:00
Ethan 3c3708f562 test(codersdk/toolsdk): cover start without auto-bump (#24918)
Previously, the `CreateWorkspaceBuild` toolsdk tests only exercised a
start where the workspace's prior template version was also the
template's active version, so they did not prove that a plain start
keeps using the previously built version.

Replace that tautological coverage with an isolated fixture that
advances the template's active version and asserts a start without
`TemplateVersionID` still reuses the prior build's version.
2026-05-07 12:41:02 +10:00
04cc983833 fix: add preset support to MCP tools (#24694)
The chat tools (`read_template`, `create_workspace`) did not surface or
respect template version presets. Presets were invisible to the LLM and
preset parameter defaults were never applied at workspace creation. The
`toolsdk` MCP surface had the same gap (ref #24695, now subsumed here).

## What this changes

- **`read_template`** returns presets with `id`, `name`, `default`,
`description`, `icon`, `parameters`, and `desired_prebuild_instances`
(when set), so the LLM can pick the right preset and prefer
prebuilt-backed ones.
- **`create_workspace`** accepts a `preset_id`. The wsbuilder applies
preset parameter defaults and may claim a prebuilt workspace.
- **`start_workspace`** does *not* accept a preset. Presets are a
creation-time choice; subsequent starts use the workspace's existing
version and parameters. Users who need a specific preset or version on
an existing chat can create the workspace out-of-band (CLI / UI / API)
with the desired configuration and attach the chat to it.
- **`toolsdk`** gains `GetTemplate` (with presets including
`desired_prebuild_instances`), preset support on `CreateWorkspace`, and
preset + `rich_parameters` support on `CreateWorkspaceBuild`. The
`template_version_preset_id` description warns about preset/version
affinity.


> 🤖 Generated with [Coder Agents](https://coder.com/agents) and reviewed by a human.

Co-authored-by: Max schwenk <maschwenk@gmail.com>
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-01 10:57:52 +01:00
Cian Johnston d5a5be116d fix: fall back to name lookup for UUID-shaped workspace names (#24340)
`namedWorkspace` in `cli/root.go` parsed workspace identifiers with
`uuid.Parse` first and returned immediately on success, even when no
workspace had that UUID as its actual ID. This caused 404 errors for any
workspace whose name was a valid 32-char hex string (dashless UUID).

- Add `codersdk.ResolveWorkspace`: tries UUID lookup first, falls back
to name lookup on 404. `NameValid` guard skips the fallback for standard
dashed UUIDs (36 chars > 32-char name limit).
- Export `codersdk.SplitWorkspaceIdentifier`, replacing the duplicate
`splitNamedWorkspace` in `cli/root.go` (uses `strings.Cut`).
- Delete `namedWorkspace` from `cli/root.go`; all 28 call sites now use
`client.ResolveWorkspace` directly.
- Delete `namedWorkspace` and `splitNameAndOwner` from
`codersdk/toolsdk/bash.go`; inline `client.ResolveWorkspace`.
- Simplify `GetWorkspace` tool handler to a single `ResolveWorkspace`
call.
- Unit tests via httptest mock cover UUID, name, owner/name, UUID-like
fallback, not-found, server error, transport error, and invalid
identifier paths.
- Integration tests in `cli/show_test.go` and `codersdk/toolsdk` for
workspaces with UUID-like names.

> Generated with Coder Agents
2026-04-27 12:58:26 +01:00
Ethan 181e103201 fix: reuse shared tailnet for coderd-hosted MCP workspace tools (#24460)
## Problem

Coderd can expose an MCP server at `/api/experimental/mcp/http` (we have
this enabled on dogfood). Its workspace tools dialed agents through a
per-call client-side tailnet stack. Every tool call re-created a
WireGuard device, netstack, magicsock + UDP sockets, DERP connection,
coordinator websocket, and their goroutines — in a process that already
runs a long-lived shared tailnet. The duplicate stacks drove up resource
usage under load.

## Fix

Route this server's tool calls through the existing shared tailnet, so
none of those transports are reconstructed per call. Closing an
`AgentConn` now releases a tunnel reference instead of tearing down a
transport.

## Potential follow-up

`coder exp mcp server` still builds a fresh tailnet per call. It pays
per-call latency and causes coordinator/DERP churn. A shared CLI tailnet
is more involved — unlike coderd, the CLI has no existing shared tailnet
to reuse, so it would need a new long-lived client-side tailnet with
reconnect, sleep/wake, and idle-destination handling. There's less
motivation to optimize this, given the client-side MCP does not compete
for resources with coderd.

Closes CODAGT-199

> Generated by mux, but reviewed by a human
2026-04-21 11:37:10 +10:00
Mathias Fredriksson 6b0bb02e5d fix: server-side diffs and stricter fuzzy splicing for edit_files (#24454)
Fixes three classes of edit_files bugs and adds structured per-file
diff output for tool callers:

- New IncludeDiff flag on FileEditRequest; when set, the agent
  returns FileEditResponse.Files[]{Path, Diff} with unified diffs
  computed via go-udiff v0.4.1 Lines + ToUnified (not Unified,
  which calls log.Fatalf on internal error).
- Fuzzy match comparators split each line into leading whitespace,
  body, trailing whitespace, and ending. The splice substitutes at
  each position: on agreement between search and replace the file's
  bytes win; on disagreement the replacement's bytes are spliced
  verbatim. Carve-outs for empty-body lines, multi-line EOF splices,
  and level-aware indent translation for inserted lines.
- Indent-unit detection (GCD for spaces, tab-priority) lets a 4sp
  LLM search insert correctly into tab or 2sp files. Falls back to
  the previous cLead-inheritance path when units can't be detected
  cleanly.
- Empty search is rejected with "search string must not be empty".
- Duplicate file paths in one request are rejected; symlink aliases
  resolved via api.resolvePath before the dedup check.
- Frontend EditFilesRenderer consumes the structured files array by
  explicit path (no label munging) with per-file synthetic fallback
  for older agents or mismatched paths. On error, no diff is
  rendered so the synthetic fallback doesn't misrepresent a
  rejected edit as applied.

Breaking change: AgentConn.EditFiles changes from (ctx, req) error
to (ctx, req) (FileEditResponse, error) in codersdk/workspacesdk.
Source-breaking for external Go consumers; no compat shim per plan
owner.

Out of scope (tracked in CODAGT-214): level-aware indent for
middle-substituted splice lines. Locked in
TestEditFiles_FuzzyIndent_InsertionLevelAware's Lock_* cases plus
TestEditFiles_ReplaceAll_FuzzyIndentGap.
2026-04-18 16:39:34 +03:00
Atif Ali bd5b62c976 feat: expose MCP tool annotations for tool grouping (#23195)
## Summary
- add shared MCP annotation metadata to toolsdk tools
- emit MCP tool annotations from both coderd and CLI MCP servers
- cover annotation serialization in toolsdk, coderd MCP e2e, and CLI MCP
tests

## Why
- Coder already exposed MCP tools, but it did not populate MCP tool
annotation hints (`readOnlyHint`, `destructiveHint`, `idempotentHint`,
`openWorldHint`).
- Hosts such as Claude Desktop use those hints to classify and group
tools, so without them Coder tools can get lumped together.
- This change adds a shared annotation source in `toolsdk` and has both
MCP servers emit those hints through `mcp.Tool.Annotations`, avoiding
drift between local and remote MCP implementations.

## Testing
- Tested locally on Cladue Desktop and the tools are categorized
correctly.

<table>
<tr>
 <td> Before
 <td> After
<tr>
<td> <img width="613" height="183" alt="image"
src="https://github.com/user-attachments/assets/29d2e3fb-53bc-4ea7-bdb3-f10df4ef996b"
/>
<td> <img width="600" height="457" alt="image"
src="https://github.com/user-attachments/assets/cc384036-c9a7-4db9-9400-43ad51920ff5"
/>
</table>

Note: Done using Coder Agents, reviewed and tested by human locally
2026-03-18 10:21:45 +00:00
Kyle Carberry 32a894d4a7 fix: error on ambiguous matches in edit_files tool (#23125)
## Problem

The `edit_files` tool used `strings.ReplaceAll` for exact substring
matches, silently replacing **every** occurrence. When an LLM's search
string wasn't unique in the file, this caused unintended edits. Fuzzy
matches (passes 2 and 3) only replaced the first occurrence, creating
inconsistent behavior. Zero matches were also silently ignored.

## Investigation

Investigated how **coder/mux** and **openai/codex** handle this:

| Tool | Multiple matches | No match | Flag |
|---|---|---|---|
| **coder/mux** `file_edit_replace_string` | Error (default
`replace_count=1`) | Error | `replace_count` (int, default 1, -1=all) |
| **openai/codex** `apply_patch` | Uses first match after cursor
(structural disambiguation via context lines + `@@` markers) | Error |
None (different paradigm) |
| **coder/coder** `edit_files` (before) | Exact: replaces all. Fuzzy:
replaces first. | Silent success | None |

## Solution

Adopted the mux approach (error on ambiguity) with a simpler
`replace_all: bool` instead of `replace_count: int`:

- **Default (`replace_all: false`)**: search string must match exactly
once. Multiple matches → error with guidance: *"search string matches N
occurrences. Include more surrounding context to make the match unique,
or set replace_all to true"*
- **`replace_all: true`**: replaces all occurrences (opt-in for
intentional bulk operations like variable renames)
- **Zero matches**: now returns an error instead of silently succeeding

Chose `bool` over `int` count because:
1. LLMs are bad at counting occurrences
2. The real intent is binary (one specific spot vs. all occurrences)
3. Simpler error recovery loop for the LLM

## Changes

| File | Change |
|---|---|
| `codersdk/workspacesdk/agentconn.go` | Add `ReplaceAll bool` to
`FileEdit` struct |
| `agent/agentfiles/files.go` | Count matches before replacing; error if
>1 and not opted in; error on zero matches; add `countLineMatches`
helper |
| `codersdk/toolsdk/toolsdk.go` | Expose `replace_all` in tool schema
with description |
| `agent/agentfiles/files_test.go` | Update existing tests, add
`EditEditAmbiguous`, `EditEditReplaceAll`, `NoMatchErrors`,
`AmbiguousExactMatch`, `ReplaceAllExact` |
2026-03-16 16:17:33 +00:00
Mathias Fredriksson 947b390c5a fix: allow agent-reported final states, add SSE reconnection (#22286)
When AgentAPI is configured, `WithTaskReporter` unconditionally
overrides all self-reported states to `working`. The intent was to
distrust the agent's `idle` and rely on the screen watcher, but the
override also blocks `failure` and `complete`, which only the agent can
produce (the screen watcher only knows `running`/`stable`). Tasks get
stuck as `working` or `null` forever.

Now only `idle` is overridden to `working`; `failure`, `complete`, and
`working` pass through as-is.

Also:

- Remove misplaced unconditional `"Failed to watch screen events"` log
that fired on every startup
- Add SSE reconnection with exponential backoff (1s-30s) in
`startWatcher` so it recovers from dropped connections instead of dying
silently
- Add `complete` to the `coder_report_task` tool enum, which the
`coder/claude-code` registry module already instructs agents to use but
was missing from the schema

Refs coder/internal#1350
2026-02-24 20:28:50 +02:00
Spike Curtis bddb808b25 chore: arrange imports in a standard way (#21452)
Fixes all our Go file imports to match the preferred spec that we've _mostly_ been using. For example:

```
import (
	"context"
	"time"

	"github.com/prometheus/client_golang/prometheus"
	"golang.org/x/xerrors"
	"gopkg.in/natefinch/lumberjack.v2"

	"cdr.dev/slog/v3"
	"github.com/coder/coder/v2/codersdk/agentsdk"
	"github.com/coder/serpent"
)
```

3 groups: standard library, 3rd partly libs, Coder libs.

This PR makes the change across the codebase. The PR in the stack above modifies our formatting to maintain this state of affairs, and is a separate PR so it's possible to review that one in detail.
2026-01-08 15:24:11 +04:00
Mathias Fredriksson dac822b7f4 refactor: remove deprecated AITaskPromptParameterName constant (#21023)
This removes the deprecated AITaskPromptParameterName constant and all
backward compatibility code that was added for v2.28.

- Remove AITaskPromptParameterName constant from codersdk/aitasks.go
- Remove backward compatibility code in coderd/aitasks.go that populated
  the "AI Prompt" parameter for templates that defined it
- Remove the backward compatibility test (OK AIPromptBackCompat)
- Update dbfake to no longer set the AI Prompt parameter
- Remove AITaskPromptParameterName from frontend TypeScript types
- Remove preset prompt read-only feature from TaskPrompt component
- Update docs to reflect that pre-2.28 definition is no longer supported

Task prompts are now exclusively stored in the tasks.prompt database
column, as introduced in the migration that added the tasks table.
2025-12-16 15:14:59 +00:00
Steven Masley 3194bcfc9e chore: distinct operations for provisioner's 'parse', 'init', 'plan', 'apply', 'graph' (#21064)
Provisioner steps broken into smaller granular actions.
Changes:
- `ExtractArchive` moved to `init` request (was in `configure`)
- Writing `tfstate` moved to `plan` (was in `configure`)
- Moved most plan/apply outputs to `GraphComplete`
2025-12-15 11:26:41 -06:00
Danielle Maywood b255827a52 chore: promote tasks to stable from experimental (#20921)
- Promote tasks from `/api/experimental` to `/api/v2`.
- Move sdk from `ExperimentalClient` to `Client`.
- Update swagger
2025-11-25 15:24:25 +00:00
Mathias Fredriksson db76541522 test(codersdk/toolsdk): fix task status race for send/logs (#20682)
Fixes coder/internal#1111
2025-11-07 12:36:09 +02:00
Mathias Fredriksson daad93967a fix(coderd): fix template ai task check error message (#20651)
Create task was still mentioning magic prompt parameter when checking
template task validity. This change updates it to only mention validity
of `coder_ai_task` resource.
2025-11-03 12:54:43 +00:00
Asher 77e2521fa0 feat: support workspace name in get workspace tool (#20474)
This lets the LLM skip the list workspace step in some cases.

Closes https://github.com/coder/internal/issues/1022
2025-10-31 15:10:36 -08:00