opensource/coder
3
0
Fork 0
mirror of https://github.com/coder/coder.git synced 2026-09-22 05:05:20 +08:00
Code Issues Packages Projects Releases Wiki Activity
15,408 Commits 479 Branches 387 Tags
02fd1cc691f42cda7edad142a9c8aba9c48a6c38
Commit Graph
2622 Commits
Author SHA1 Message Date
Michael Suchacz 02fd1cc691 feat: allow spawn_agent model and reasoning effort override (#27385) 2026-07-22 18:43:22 +02:00
Sas SwartandClaude Opus 4.8 f17d488479 feat: add network call badges to AI sessions table (#27341) 
Surface the total and blocked Agent Firewall network calls on the AI
sessions list. Sessions that did not pass through Agent Firewall show as
"Disabled".

<img width="2842" height="1366" alt="image"
src="https://github.com/user-attachments/assets/2a68b4a9-4d93-454d-a06e-5f0d0b734a33"
/>

---------

Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
 2026-07-22 14:16:31 +02:00
Nick Vigilante f5e0c1a860 fix: correct invalid inline HTML in hand-written docs (#27298) 
## What

Fixes three classes of invalid inline HTML in hand-written docs, all of
which
render incorrectly (or only render by accident) today. Found via a
systematic,
markdown-aware audit of every `.md` under `docs/` (ignores code blocks,
inline
code, comments, and autolinks), so this is a complete sweep of the
hand-written
surface, not a spot fix.

## Changes

1. **`<kdb>` → `<kbd>` (72 tags).** The keyboard element is `<kbd>`;
`<kdb>` is
a typo that is not a real element, so renderers drop/mangle it and the
   keystrokes lose their styling. Corrected across the IDE access guides
(`cursor.md`, `windsurf.md`, `antigravity.md`). The correct `<kbd>` is
   already used in the JetBrains Gateway guide.
2. **Unclosed `<div class="tabs">` in `docs/admin/users/idp-sync.md`.**
The
"Provider-Specific Guides" section opened a `.tabs` container (rendered
as
the `DocsTabs` component) that was never closed, so the wrapper leaked
over
the rest of the page. Added the missing `</div>` before `## Next Steps`,
   matching the three other tab sections in the same file.
3. **`<Image>` → `<img>` (6 tags).** `<Image>` is not a registered docs
component — it renders only because the HTML5 parser rewrites the legacy
`<image>` tag to `<img>`. Converted to lowercase `<img>` for correctness
and
   clarity; rendering is unchanged. (`organizations.md`, `idp-sync.md`,
   `add-envbuilder.md`.)

## Scope / what is intentionally not here

- **Generated reference docs.** The audit also found swallowed
placeholders in
  generated pages (`<server>` in `reference/api/{chats,schemas}.md`;
`<glob>`/`<host>` in `agent-firewall`; `<region>` in `server`). Those
are
fixed at the generator source (codersdk comments / CLI flag help) and
tracked
  in DOCS-551.
- **`<b>Resource<b>`** in the generated audit-logs table was fixed
separately in
  #27293 (merged) and is not duplicated here.
- **`<children></children>`** is an intentional, renderer-implemented
docs
component (child-page card grid) with no HTML equivalent, so it is left
as-is.
It is well-formed; a follow-up CI checker will still verify its
open/close
  balance.

A follow-up adds CI enforcement so invalid inline HTML can't regress.

<details>
<summary>Verification</summary>

Run against the changed files:

- `markdownlint-cli2` — 0 errors
- `markdown-table-formatter --check` — no changes needed
- `typos --config .github/workflows/typos.toml` — clean
- Re-running the audit scanner: hand-written `unclosed`, `<kdb>`, and
  capitalized-component findings all drop to 0 (only the generated-doc
  placeholders tracked in DOCS-551 remain).

</details>

## Linear

DOCS-581:
https://linear.app/codercom/issue/DOCS-581/audit-and-fix-all-invalid-html-across-the-docs

> This PR was created with AI assistance (Coder Agents).
 2026-07-21 19:59:59 +00:00
Nick Vigilante f55be09bfc docs(docs/ai-coder/ai-gateway): fix bmcp_ described as suffix instead of prefix (#27392) 
The Tool Injection section of the AI Gateway MCP doc called `bmcp_` a
suffix, directly contradicting the correct description one section
earlier on the same page and the `aibridge` implementation, where
`injectedToolPrefix` is prepended to every bridged MCP tool name
(`aibridge/mcp/tool.go`).

Reported by a customer who read the suffix wording and assumed `bmcp` in
a tool name like `bmcp_github_list_gists` was a typo.

---
🤖 Built with AI assistance.
 2026-07-21 11:59:30 -05:00
Atif Ali 2b2a5c963a Revert "fix(coderd): explain default GitHub app org visibility on login rejection" (#27388) 2026-07-21 16:17:27 +00:00
Atif Ali 48e9bb3391 fix(coderd): explain default GitHub app org visibility on login rejection (#27374) 
## Problem

On a fresh deployment with no custom GitHub OAuth app, Coder falls back
to the default Coder-managed GitHub app. That app can only see
organization memberships in organizations where it has been installed.
If `CODER_OAUTH2_GITHUB_ALLOWED_ORGS` is set but the app isn't installed
in the allowed organizations, the membership list comes back empty and
every login, including the first admin login, is rejected with a bare
"You aren't a member of the authorized Github organizations!" with no
hint about the actual cause. This leaves fresh deployments in an
apparently broken state.

## Fix

* Append a remediation hint to the login rejection when the default
provider is configured, pointing at the [app installation
page](<https://github.com/apps/coder/installations/select_target>) and
at configuring a custom GitHub OAuth app.
* Log a startup warning when the default provider is combined with
`CODER_OAUTH2_GITHUB_ALLOWED_ORGS`, listing the allowed orgs and the
install URL.
* Document the installation requirement next to the
`CODER_OAUTH2_GITHUB_ALLOWED_ORGS` step in the GitHub auth docs.

Access-control behavior is unchanged; the org check still rejects logins
as before, it just explains why and how to fix it.

## Testing

* New `TestUserOAuth2Github/NotInAllowedOrganizationDefaultProvider`
asserts the hint appears when `DefaultProviderConfigured` is set; the
existing `NotInAllowedOrganization` subtest asserts it does not leak
into the custom-app path.

Fixes coder/coder#17752
 2026-07-21 20:43:10 +05:00
Sas SwartandClaude Opus 4.8 a9a1dcc65d feat: add network calls column to AI sessions table (#27269) 
Add a "Total/blocked network calls" column to the AIBridge sessions
table.

Update `ListAIBridgeSessions` query to calculate network called made and
blocked per session. See query plan
[here](https://explain.dalibo.com/plan/54355c90b165ggb4).

---------

Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
 2026-07-21 14:34:23 +02:00
Michael Suchacz 3227cac217 feat: add manual chat compaction via /compact (#27081) 
Adds a user-triggered `/compact` action for Coder Agents chats: typing
`/compact` in the composer (or picking it from the `/` trigger menu)
summarizes the conversation so far to free up context window space.

## How it works

- New `POST /api/experimental/chats/{chat}/compact` endpoint
(owner-only, RBAC `ActionUpdate`, excluded from the public API reference
via `x-apidocgen skip`). It marks the chat with a durable one-shot
`chats.compaction_requested_at` signal and moves it `waiting -> running`
via a new `RequestCompaction` state transition; no message row is
inserted. AI Gateway attribution needs no per-request key: generation
preparation resolves the owner's synthetic API key (#27170) like any
other turn.
- `RequestCompaction` hands off chat ownership (clears
`worker_id`/`runner_id`) so a worker acquisition hint is published;
since the transition changes no history, the previous runner could
otherwise miss the request under reordered pubsub delivery.
- The background chat worker picks the chat up like any other turn. A
pending manual request takes precedence over turn completion in the
generation decision, and forces compaction even below the automatic
threshold (and when compaction is disabled via threshold=100). The
commit step consumes the request marker in the same transaction; any
transition that ends the turn clears stale markers.
- The summary triplet reuses the automatic-compaction path, now tagged
with a `source` (`automatic` | `manual`) that is plumbed through
streamed progress parts, persisted tool JSON, and the UI label
("Summarized (manual)").
- Validation order: busy chats reject with 409 (state-machine conflict),
empty/already-compacted chats with 409 "nothing to compact", archived
chats with 400; the owner usage-limit check runs last so no-op requests
surface the specific conflict instead of a limit error.
- Web UI: the `/` trigger menu now has a built-in "Commands" group
listing `/compact`; submit intercepts exactly `/compact` and calls the
endpoint instead of sending a message. A personal or workspace skill
named `compact` takes precedence over the built-in command; while skill
collisions are still resolving, an exact `/compact` submission is
blocked with a retryable hint instead of leaking as message text.
History and queued-message edits are never intercepted. After
compaction, the context usage indicator resets to its unknown state
until the next assistant response reports fresh usage, instead of
showing the stale pre-compaction number.
- codersdk: `ExperimentalClient.CompactChat`.

Worker-path execution (rather than compacting synchronously in the
handler) reuses the existing lock fencing, live "Summarizing..."
streaming, retry accounting, restart resilience, and debug-run
observability. Rationale documented in `coderd/x/chatd/ARCHITECTURE.md`.

## Testing

- State machine: transition-matrix coverage for `RequestCompaction`,
marker lifecycle tests (carried by lease renewals/queue appends, cleared
by terminal transitions, consumed by commit), ownership handoff +
acquisition hint assertions.
- Worker: decision-ordering and forced-compaction unit tests;
active-server end-to-end test (manual compact below threshold produces a
`source=manual` summary, returns to `waiting`, no assistant follow-up;
busy chat rejected).
- API: success, archived, non-owner, RBAC-denied, empty-chat, no-daemon
cases; usage-limit ordering (at-limit owners still get
state/nothing-to-compact conflicts for no-op requests, with marker
rollback).
- Frontend: Storybook play tests for the Commands menu group, submit
intercept, skill-name collision, queued-edit passthrough, and
manual/automatic tool rendering; unit tests for command availability
resolution and the post-compaction context usage reset.

> This PR was created by Mux, an AI coding agent, working on Mike's
behalf.
 2026-07-21 10:58:08 +02:00
Susana Ferreira a9fdf87a2f feat: add GET /groups/{group}/members/ai/spend (#27130) 
## Description

Adds `GET /api/v2/groups/{group}/members/ai/spend?user_ids=...` (also available org-scoped at `/api/v2/organizations/{org}/groups/{groupName}/members/ai/spend`) to return per-member AI spend attributed to a group, along with each member's effective budget group and the applied spend limit when the queried group is their effective budget source.

In the UI, this endpoint is used alongside the existing `/api/v2/groups/{group}/members` endpoint. AI spend data is kept separate from that endpoint so that:

- Different concepts stay on different endpoints: identity (group members) vs. cost control (spend). Cost control is an additional feature layered on top of groups/orgs.
- Callers that don't need spend information don't pay for its computation.

UI flow:

1. Request `/api/v2/groups/{group}/members` → returns the group's members.
2. Request `/api/v2/groups/{group}/members/ai/spend?user_ids=...` with the IDs from step 1.

**Note:** Only current members of the queried group are returned. `spend_limit_micros` and `limit_source` are populated only when the queried group is the member's effective budget source (its own limit or a user override). `effective_group_id` is null when the member's budget resolves to a group in another organization, since an organization is treated as a tenant boundary.

<img width="2880" height="1904" alt="image" src="https://github.com/user-attachments/assets/33ed395d-d1a3-4b46-bb04-c8d3f41c8886" />

## Changes

- Add `codersdk.GroupMembersAISpend` and `GroupMemberAISpend` types, reusing the shared `AISpendPeriodWindow`.
- Add `GetGroupMembersAISpend` SQL query with a dbauthz per-row filter that mirrors `GET /api/v2/groups/{group}/members`.
- Add handler and routes under `/groups/{group}/members/ai/spend` (and the org-scoped alias) with a required `user_ids` query param (cap 100). Callers with more than 100 members are expected to batch across multiple requests.
- Add codersdk client method.
- Tests: dbauthz, raw SQL, endpoint, and role-access.

Closes https://linear.app/codercom/issue/AIGOV-471/backend-group-members-endpoint-with-members-spend

> [!NOTE]
> Initially generated by Claude Opus 4.7, modified and reviewed by @ssncferreira
 2026-07-20 13:04:37 +01:00
Susana Ferreira 2adc8f5272 feat: add GET /organizations/{org}/groups/ai/spend (#27123) 
## Description

Adds `GET /api/v2/organizations/{org}/groups/ai/spend?group_ids=...` to return per-group AI spend and configured limits for a set of groups in an organization.

In the UI, this endpoint is used alongside the existing `/api/v2/organizations/{org}/groups` endpoint. AI spend data is kept separate from that endpoint so that:

- Different concepts stay on different endpoints: identity (groups) vs. cost control (spend). Cost control is an additional feature layered on top of groups/orgs.
- Callers that don't need spend information don't pay for its computation.

UI flow:

1. Request `/api/v2/organizations/{org}/groups` → returns the organization's groups.
2. Request `/api/v2/organizations/{org}/groups/ai/spend?group_ids=...` with the IDs from step 1.

The groups endpoint from 1) is currently not paginated, but if pagination is added later, this design keeps the two responses in sync. This spend endpoint intentionally takes `group_ids` rather than paginating on its own, since it depends on the group set from step 1. Pagination could be added in the future, especially for Cost Control-focused pages.

<img width="2880" height="1460" alt="image" src="https://github.com/user-attachments/assets/ea83b74d-6a4f-45a6-af2f-1024e019da07" />

## Changes

- Add `codersdk.OrganizationGroupsAISpend` and `OrganizationGroupAISpend` types, plus a shared `AISpendPeriodWindow` embedded in the spend response.
- Add `GetOrganizationGroupsAISpend` SQL query with a dbauthz per-row filter that mirrors `GET /organizations/{org}/groups`.
- Add handler and route under `/organizations/{organization}/groups/ai/spend` with a required `group_ids` query param (cap 100). Callers with more than 100 groups are expected to batch across multiple requests.
- Add codersdk client method.
- Tests: dbauthz, raw SQL, endpoint, and role-access.

Closes https://linear.app/codercom/issue/AIGOV-466/backend-organization-groups-endpoint-with-groups-spend

> [!NOTE]
> Initially generated by Claude Opus 4.7, modified and reviewed by @ssncferreira
 2026-07-20 12:54:52 +01:00
Michael Suchacz 9f4ddea571 feat: revoke MCP server OAuth grants at the provider on disconnect (#27300) 
Closes
[CODAGT-805](https://linear.app/codercom/issue/CODAGT-805/revoke-oauth-grants-at-the-source-for-mcp-servers).

The experimental MCP server OAuth2 disconnect endpoint previously
deleted only the stored token row, leaving the grant active at the OAuth
provider. This PR adds provider-side token revocation while keeping
local disconnect independent of provider availability.

## Changes

- Add `mcp_server_configs.oauth2_revocation_url` in migration `000547`.
The value can be configured manually, discovered from RFC 8414 metadata,
and managed through the MCP server settings UI. Non-admin responses
redact it with the other OAuth2 fields.
- Revoke the refresh token first through the RFC 7009 endpoint, then
fall back to the access token only for `unsupported_token_type`. Public
clients send `client_id`; confidential clients use
`client_secret_basic`.
- Delete the local token transactionally before best-effort provider
revocation. Callers without a token receive the same response for hidden
and nonexistent config IDs, and provider failures return a generic
warning without exposing provider response bodies.
- Require HTTPS revocation endpoints except for HTTP loopback URLs.
Redirects must preserve the POST and remain on the configured origin.
Redirect errors omit provider-controlled paths and query strings so
reflected token material cannot enter logs.
- Treat `200 OK` and `204 No Content` as completed revocations. `202
Accepted` remains a failure because it does not confirm completion.
- Prevent an in-flight refresh from recreating a token deleted by
disconnect. Refresh persistence now uses an optimistic update keyed by
token ID and `updated_at`; only the OAuth callback can create a token
row. Refresh conflicts reload the current row or clear in-memory auth
when disconnect deleted it.
- Return `{token_revoked, token_revocation_error}` from disconnect,
while retaining SDK compatibility with the legacy `204` response. The UI
surfaces provider revocation failures as warning toasts.
- Document revocation endpoint discovery, HTTPS requirements, and
best-effort disconnect behavior.

No token or no configured revocation URL returns `token_revoked: false`
without an error, so disconnect remains idempotent.

> Updated by Mux, an AI coding agent, on Mike's behalf.
 2026-07-20 00:14:03 +02:00
Nick Vigilante 4d4d2575e4 fix: exempt docs/.style/ prose from Vale's own Coder rules (#27192) 
## Problem

The repo-root `.vale.ini` lints all of `docs/**/*.md`, including the
style guide under `docs/.style/style-guide/`. The guide deliberately
contains the constructs the Coder rules ban:

- Don't examples in blockquotes and in the Don't column of Do/Don't
tables.
- Banned terms named in prose and headings (a rule about "Next steps"
must write "Next steps").

`Coder.BrandNames` already fired on the Brand names table's Don't column
(`Hashicorp`, `HASHICORP`), and each future `Coder.*` rule would fire on
its matching section's examples. Vale runs advisory, so these surface as
non-blocking annotations, exactly the standing backlog the README warns
against, and they break the documented "zero baseline findings"
doctrine.

## Change

Option B, narrowed after review to the style-guide subtree:

- `.vale.ini`: clear `BasedOnStyles` for `docs/.style/style-guide/**`
(the guide subpages only). The annotation demo and the rest of
`docs/.style/` stay linted; the demo keeps firing its `Coder.Demo*`
rules by inheriting `Coder` from `[*.md]`, with no re-include. Also
corrects the `error`-tier doctrine line to reflect advisory-only Vale.
- `docs/.style/README.md`: "What still runs" / "What does not run"
describe the narrowed exemption, and record that it holds only for paths
that resolve to `docs/.style/style-guide/...` (repo-root-relative), not
absolute or subdirectory-relative invocations.
- `docs/.style/style-guide/README.md`: exemption note scoped to
`docs/.style/style-guide/**`.
- `docs/.style/styles/Coder/BrandNames.yml`,
`docs/.style/styles/Coder/README.md`: fix the remaining dead
`style-guide.md` links (the `BrandNames.yml` one was a live 404 surfaced
in Vale annotations).

## Validation

- `make lint/prose` reports only the 3 intended `Coder.Demo*` findings
across the corpus; the guide's `BrandNames` findings are gone.
- `content-guidelines.md`, the landing `README.md`, and
`styles/Coder/README.md` are now linted and clean (0). Injecting a
miscased `Hashicorp` into `content-guidelines.md` fires
`Coder.BrandNames`; reverted.
- The guide subpages stay exempt (0); the annotation demo still fires
its 3 findings via `[*.md]`.
- Invocation-path behavior verified against
`docs/.style/style-guide/word-choice.md`: repo-root-relative `docs/...`
is exempt (0), while `docs/`-relative and absolute-path invocations
still fire 2 `Coder.BrandNames` errors.
- `markdownlint-cli2` clean on the changed READMEs.

## Decision log

Considered three options from the issue:

- **A. Surgical in-file suppression** (`vale off`/`vale on` comment
blocks, backticking banned terms). Rejected as primary: noisy in source,
repeated per example, bare-term headings still fire, and every new rule
can re-break the guide. It is path-independent, so it would have covered
the editor case noted in Option B's limitation.
- **B. Config-level subtree exemption** (chosen; narrowed to the guide
subtree after round-2 review). One place, zero source noise,
future-proof for the guide subpages.
- **Scope.** The clear applies to `docs/.style/style-guide/**` only.
Round-2 review (three reviewers converged, verified end-to-end) showed
the original directory-wide clear over-reached: `content-guidelines.md`
and the landing `README.md` are ordinary contributor prose that lints
clean today (`content-guidelines.md` uses `HashiCorp` correctly), so
they should stay linted. Narrowing to the guide subtree preserves the
exact 3-finding baseline, relints those two pages, removes the
last-match-wins ordering hazard, and drops the whole-package demo
re-include. The demo sits at the `.style/` root and keeps firing via
`[*.md]`.
- **Tradeoff.** The guide's own prose under `style-guide/` is no longer
auto-linted by Vale. Acceptable because the guide must contain banned
terms by design, one-sentence-per-line is enforced by markdownlint, and
human review covers the rest.
- **Limitation.** The exemption holds only for invocations whose path
resolves to `docs/.style/style-guide/...` (repo-root-relative
`docs/...`, what `make lint/prose` and CI pass). Absolute paths
(editors/LSP) and subdirectory-relative paths still flag the guide.
Recorded in `docs/.style/README.md`.
- **C. Restructure examples into fenced code blocks.** Rejected: changes
the guide's rendered format and doesn't cover Do/Don't tables or
headings.

## Notes

- `.vale.ini` is outside `docs/`, so this PR modifies repo-wide lint
config; the clear is scoped to `docs/.style/style-guide/` only.
- Linear: DOCS-543.

---
Opened via Coder Agents on behalf of @nickvigilante.
 2026-07-18 13:41:27 -04:00
Michael Suchacz 1ac106255b feat: add Anthropic 1M context window toggle for Agents model configs (#27257) 2026-07-17 13:04:13 +02:00
Yevhenii Shcherbina 1453b4b1fc docs: document Bedrock mantle protocol (#27296) 
Implements:
https://linear.app/codercom/issue/AIGOV-518/add-docs-for-bedrock-mantle
Follow-up to: https://github.com/coder/coder/pull/26745
 2026-07-16 14:58:09 -04:00
Nick Vigilante 77582be805 fix: close <b> tag in generated audit log table header (#27293) 
## What

The audit log resource table header in
`docs/admin/security/audit-logs.md` was
emitted as `<b>Resource<b>`: a second opening `<b>` instead of a closing
`</b>`.
Because the bold element never closes, Markdown/HTML renderers can bold
content
well beyond the header cell.

The page is generated (`<!-- Code generated by 'make
docs/admin/security/audit-logs.md'. DO NOT EDIT -->`),
so the fix belongs in the generator, `scripts/auditdocgen/main.go`, with
the
doc regenerated from it.

## Changes

- `scripts/auditdocgen/main.go`: emit a closing `</b>` instead of a
second `<b>` in the table header row.
- `docs/admin/security/audit-logs.md`: regenerated with `make
docs/admin/security/audit-logs.md`; only the header cell changes.

## Verification

<details>
<summary>Regenerated doc and local checks</summary>

Header cell before (unclosed tag):

```text
| <b>Resource<b>                                   | ...
```

Header cell after (balanced tag):

```text
| <b>Resource</b>                                  | ...
```

- `make docs/admin/security/audit-logs.md` regenerates the page from the
fixed generator and changes only the header cell (single line; the table
stays aligned).
- Local `make pre-commit` passed with `GEN_SKIP_GOLDEN=1` (this
workspace has no Docker daemon for the golden-file gen step, which this
change does not touch): `gen`, `fmt`, `lint/go`, `lint/ts`,
`lint/markdown`, `lint/typos`, `lint/emdash`, and the slim binary build
all green.

</details>

## Linear

DOCS-580:
https://linear.app/codercom/issue/DOCS-580/fix-unclosed-b-tag-in-generated-audit-logs-table-header

---

This PR was created using AI (Coder Agents) on behalf of @nickvigilante,
who is accountable for its contents. See the [AI Contribution
guidelines](https://coder.com/docs/about/contributing/AI_CONTRIBUTING).
 2026-07-16 14:56:55 +00:00
Cian Johnston f7481c5d08 feat: Add full text search over chat messages (#27126) 
Closes CODAGT-721
Closes CODAGT-722
Closes CODAGT-723
Closes CODAGT-724
Closes CODAGT-725

This PR adds the database and API pieces necessary to support full-text
chat message search.

- Adds required chat schema for full-text search
- Adds dbpurge job to populate search_tsv in the background
- Adds `search` parameter to GetChats query
- Adds `search` filter to `searchquery.Chats`
- Wires chat search filter into chats API

> Implemented by Coder Agents, reviewed and tested by a human.
 2026-07-16 15:21:57 +01:00
Ehab Younes 35ade9e3d2 feat: collect workspace logs in support bundles (#26694) 
Add workspace-side file collection to `coder support bundle` via
repeatable --workspace-file flags. The agent resolves the requested
paths or globs inside the remote workspace and streams back a tar with
a manifest and the collected files; nothing is read from the machine
running the command.

- Add POST /api/v0/bundle-files to the agent's agentfiles package.
- Expand env vars in the agent's environment; paths must then be
  absolute or start with ~/ (the agent user's home directory).
- Support ** globs and tail oversized files.
- Record requested patterns, per-path errors, truncation, and the
  applied limits in a manifest.
- Unpack the archive into the bundle under agent/workspace_files/,
  recording dropped entries in collection_errors.txt.
- Write a manifest-only archive marking collection as unsupported for
  agents that predate the endpoint.
- Bound collection: 64 KB request body, 10000 files, 10 MiB per file,
  100 MiB total including archive overhead, 110 MiB client-side read
  cap, 5 minute timeout.

Closes #26020
 2026-07-16 13:00:32 +03:00
Nick Vigilante c84aa564ba docs: normalize code-fence languages for Shiki compatibility (#27161) 
Normalizes non-standard code-fence language tags across `docs/**` so a
strict highlighter (Shiki, used by Fumadocs) won't fail the build on an
unrecognized language, and unifies redundant synonym tags onto one
canonical form per language. The current renderer (Speed-Highlight)
detects the language from the code content, not the fence label, so this
drift wasn't visible until now.

## Changes

- `hcl` -> `tf` (199 fences, including indented ones nested in
numbered/bulleted lists). Shiki ships `hcl` and `terraform` as two
distinct grammars (not aliases); every `hcl`-tagged fence in `docs/**`
is actually Terraform resource/data/provider syntax, so the more
specific `terraform` grammar is correct for all of them. `tf` is Shiki's
own alias for that grammar, and it's also what GitHub's own markdown
renderer resolves to the same HCL/Terraform highlighting.
- `pwsh`/`powershell` -> `ps1`. Both `ps` and `ps1` are registered
PowerShell aliases in Shiki, but on GitHub's renderer only `.ps1` is a
registered file extension (`.ps` isn't), so `ps1` renders identically to
`powershell` there today while bare `ps` would silently lose
highlighting.
- `env` -> `dotenv` (a dedicated Shiki grammar for `KEY=VALUE` files)
- `text`/`output`/`none`/`url` -> `txt`. Same built-in plain-text
fallback either way, just shorter.
- `Dockerfile` -> `dockerfile` (lowercase)
- `bash`/`shell` -> `sh` (732 fences). Shiki and GitHub both alias all
three to a single shell grammar; this was already the style guide's
stated preference, just not enforced across the existing corpus until
now.
- `markdown` -> `md` (4 fences). Alias of the same grammar in both Shiki
and GitHub.
- `jsonc` -> `json` (1 fence). The block has no comments or trailing
commas, so it doesn't need the comments-capable grammar.
- `ts` -> `tsx` (2 fences, `docs/about/contributing/frontend.md`).
Verified the actual content tokenizes identically under both grammars,
and a sibling block in the same file already needs `tsx` for real JSX,
so unifying to one tag is safe for this file. Documented a caveat: `tsx`
mis-tokenizes the legacy angle-bracket type-assertion syntax
(`<Type>value`), which is invalid in real `.tsx` files anyway, so use
`value as Type` instead.
- `yml` -> `yaml` (1 fence)
- Updated `docs/.style/style-guide/formatting.md` to document all
canonical tags

`promql` (2 fences) and `caddyfile` (2 fences) are left as-is. Shiki
doesn't bundle a grammar for either, so they need a custom grammar
registration when the site adopts Shiki, rather than degrading to `txt`.
Tracked as follow-up work under DOCS-118 and
[DOCS-544](https://linear.app/codercom/issue/DOCS-544/vendor-a-local-promql-grammar-for-shiki-syntax-highlighting)
(promql).

Does not touch `offlinedocs/`.

Linear:
[DOCS-476](https://linear.app/codercom/issue/DOCS-476/normalize-docs-code-fence-languages-de-risk-shikifumadocs)

<details>
<summary>How the fence tags were verified</summary>

Each tag was tested against a real `shiki@latest` highlighter instance
(`codeToHtml`/`codeToTokens`) and cross-checked against GitHub's
`@wooorm/starry-night` grammar sources (the renderer that actually
displays these `.md` files today, in repo browsing and PR diffs), since
that's what determines whether brevity is safe before Shiki adoption:

```text
FAIL  env        -- Language `env` is not included in this bundle.
FAIL  Dockerfile -- Language `Dockerfile` is not included in this bundle.
FAIL  promql     -- Language `promql` is not included in this bundle.
FAIL  caddyfile  -- Language `caddyfile` is not included in this bundle.
FAIL  pwsh       -- Language `pwsh` is not included in this bundle.
FAIL  output     -- Language `output` is not included in this bundle.
```

`hcl` doesn't error in Shiki, since it's a real grammar, but that's
exactly the trap: it was silently rendering every fence with the generic
HCL grammar instead of the Terraform-specific one. Every `hcl`-tagged
fence in `docs/**` was manually checked against `origin/main` and is
genuinely Terraform content.

For `ts`/`tsx`, tokenizing the actual doc content confirmed identical
output under both grammars; a synthetic test with the legacy
angle-bracket cast syntax confirmed `tsx` degrades on that specific
construct, which the style guide now calls out.

The first normalization pass only matched fence tags at column 0
(`^```tag$`), missing tags indented inside numbered/bulleted lists. A
follow-up pass caught the remaining occurrences at any indentation
level.

</details>


---

*This PR description and the underlying changes were prepared with Coder
Agents assistance.*
 2026-07-15 14:07:09 -04:00
Ethan cc11c8a536 feat: surface model content-filter refusals as a blocked chat error (#27118) 
Blocked turns from a provider's content filter (Anthropic's `refusal`
stop reason with empty content) previously ended silently on the
"Thinking" spinner. They now end as a terminal `content_filter` error
that renders as a "Response blocked" message with the provider's
category and explanation.

<img width="888" height="335" alt="image"
src="https://github.com/user-attachments/assets/cef85a59-4091-4e62-9d45-1eb06748db48"
/>


Closes CODAGT-611

Follow-ups will involve implementing fallbacks, but this alone is pretty
important
 2026-07-15 16:36:31 +10:00
Bobby Ho 0207a9824f fix: enforce max body size on CSP violation report endpoint (#27243) 
The `/api/v2/csp/reports` endpoint is unauthenticated and CSRF-exempt,
since it's the browser's `report-uri` target, and decoded request bodies
with no size limit. This let an attacker post arbitrarily large JSON
bodies to force unbounded heap allocation and OOM the server (Cure53
CDM-02-007).

Wraps the request body in `http.MaxBytesReader` before decoding and
returns 413 when the limit is exceeded, matching the existing convention
used by `files.go`, `aitasks.go`, and `exp_chats.go`.

Fixes: https://github.com/coder/security-disclosures/issues/171
 2026-07-14 12:29:38 -07:00
Nick Vigilante 61e52532c0 docs: wrap placeholder syntax in backticks in CLI help and swagger annotations (#27194) 
## Problem

Generated reference docs (`docs/reference/cli/*`,
`docs/reference/api/*`) contained raw placeholder and JSON syntax that
came straight from Go CLI help strings and swagger annotations. HTML
renderers treat the angle-bracket tokens (`<team-slug>`, `<uuid>`,
`<KEY>`, etc.) as unknown tags and drop them, so readers see
broken/half-missing text today. The same strings also break MDX parsing.

## Fix

Wrap the placeholder/JSON syntax in backticks **at the source** (Go help
strings and swagger annotation comments), then `make gen`. Rendered docs
now show the placeholders as inline code instead of dropping them.

### Source changes

| File | Placeholder wrapped | Surfaces in |
|------|--------------------|-------------|
| `codersdk/deployment.go` | `` `<organization-name>/<team-slug>` `` |
`cli/server.md`, `coder --help`, settings UI |
| `codersdk/deployment.go` | `` `CODER_AI_GATEWAY_PROVIDER_<N>_*` ``, ``
`CODER_AI_GATEWAY_PROVIDER_<N>_<KEY>` `` | `api/schemas.md` |
| `cli/tokens.go` | `` `<type>:<uuid>` `` | `cli/tokens_create.md`,
`coder --help` |
| `coderd/aitasks.go` | `` `owner:<…>` ``, `` `organization:<…>` ``, ``
`status:<status>` `` | `api/tasks.md` |
| `coderd/exp_chats.go` | `` `pr_status:<…>` `` and sibling filter
tokens | `api/chats.md` |
| `coderd/provisionerdaemons.go`, `coderd/provisionerjobs.go` | ``
`{'tag1':'value1','tag2':'value2'}` `` | `api/organizations.md`,
`api/provisioning.md` |

Everything else in the diff (`coderd/apidoc/*`, `docs/reference/**`,
`*.golden`, `site/src/api/typesGenerated.ts`) is `make gen` output.

## Reviewer notes (the "considered pass" from the ticket)

- **Product-visible:** this changes `coder server --help` and `coder
tokens create --help` output, and the `server-config.yaml` reference
comment. Backticks in terminal help are literal but read fine as
placeholder markers.
- **Settings UI:** the `deployment.go` `Description` also renders in the
deployment settings page. If that field is not Markdown-rendered,
literal backticks will show there. Happy to drop the `deployment.go`
change if you'd rather keep the UI text clean and fix `server.md`
another way.
- **Out of scope here:** `docs/reference/cli/agent-firewall.md`
(`<host>`/`<glob>`) is generated from the external
`github.com/coder/boundary` module, not this repo. It needs an upstream
fix + module bump; not included in this PR.

<details>
<summary>Implementation notes / decision log</summary>

- Scope taken from DOCS-551: source-level backtick pass for generated
reference docs only. Hand-written Markdown fixes are tracked separately
(companion ticket).
- Swagger `@Param` descriptions are Go comments, so the existing `\|`
pipe-escaping in the chats `q` filter is preserved inside the new
backticks (still required for the Markdown table cell to render `|`).
- Verified after `make gen`: generated docs render placeholders as code
spans, table pipes intact; `gofmt` clean; changed Go packages build; no
emdash/endash introduced.
- Deliberately left the `AIProviderConfig` type-level doc comment
untouched because it does not surface in any generated doc (kept the
diff to doc-feeding comments).

</details>

Linear: DOCS-551

---

_Opened by Coder Agents on behalf of @nickvigilante._



---

## Evidence: placeholders dropped on the live docs site

Verified **2026-07-14** against the live site (`coder.com/docs`, i.e.
`main`, pre-merge) by loading each affected page in headless Chrome and
reading the post-hydration DOM (confirmed identical in the raw page
payload). Each simple `<token>` placeholder is parsed as an **empty
custom HTML element**, so the browser renders nothing for it and the
placeholder text disappears from the page.

### What readers see today (before this PR)

| Page (live) | Source Markdown | Rendered on the live site |
|-------------|-----------------|---------------------------|
| [`cli/server`](https://coder.com/docs/reference/cli/server) — OAuth2
GitHub Allowed Teams | `Structured as: <organization-name>/<team-slug>.`
| `Structured as: /.` |
|
[`cli/tokens_create`](https://coder.com/docs/reference/cli/tokens_create)
— `--allow` | `Repeatable allow-list entry (<type>:<uuid>, e.g.
workspace:1234-...).` | `Repeatable allow-list entry (:, e.g.
workspace:1234-...).` |
| [`api/tasks`](https://coder.com/docs/reference/api/tasks) — `q` | `...
status:<status>` | `... status:` (nothing after the colon) |
| [`api/schemas`](https://coder.com/docs/reference/api/schemas) —
AIBridgeConfig (`anthropic`/`bedrock`/`openai`) |
`CODER_AI_GATEWAY_PROVIDER_<N>_*` | `CODER_AI_GATEWAY_PROVIDER__*` |
| [`api/schemas`](https://coder.com/docs/reference/api/schemas) —
AIBridgeConfig (`providers`) | `CODER_AI_GATEWAY_PROVIDER_<N>_<KEY>` |
`CODER_AI_GATEWAY_PROVIDER__` |

[`api/chats`](https://coder.com/docs/reference/api/chats) (`q`) drops
five tokens the same way — `title:<substring>`, `diff_url:<url>`,
`pr:<number>`, `pr_title:<text>`, and the trailing `title:<value>`. The
live parameter description reads (note the dangling `title:`,
`diff_url:`, `pr:`, `pr_title:`):

```text
Search query. Supports title: (case-insensitive, quote multi-word values), archived:bool, has_unread:bool, pr_status:<draft|open|merged|closed> as repeated or comma-separated values, source:<created_by_me|shared_with_me>, diff_url: (quote values containing colons), pr: (exact PR number match), repo:<owner/repo> (case-insensitive substring match against git remote origin or URL), pr_title: (case-insensitive PR title substring). Bare terms are not supported; use title: for title filtering.
```

<details>
<summary>Raw rendered DOM from the live site (headless Chrome,
post-hydration)</summary>

```html
<!-- reference/cli/server -->
Structured as: <organization-name>/<team-slug>.</team-slug></organization-name>

<!-- reference/cli/tokens_create -->
Repeatable allow-list entry (<type>:<uuid>, e.g. workspace:1234-...).</uuid></type>

<!-- reference/api/tasks : only status:<status> drops; the /-containing tokens are escaped and survive -->
Search query for filtering tasks. Supports: owner:&lt;username/uuid/me&gt;, organization:&lt;org-name/uuid&gt;, status:<status></status>

<!-- reference/api/schemas : anthropic / bedrock / openai rows -->
Deprecated: Use Providers with indexed CODER_AI_GATEWAY_PROVIDER_<n>_* env vars instead.</n>

<!-- reference/api/schemas : providers row -->
Providers holds provider instances populated from CODER_AI_GATEWAY_PROVIDER_<n>_<key> env vars and/or the deprecated LegacyOpenAI/LegacyAnthropic/LegacyBedrock fields above.</key></n>
```

The parser auto-inserts closing tags
(`</team-slug></organization-name>`) and lowercases the tag name (`<N>`
becomes `<n>`), leaving `__` where `<N>_` used to be. Every wrapped
placeholder renders correctly as inline code on the [docs preview for
this
branch](https://coder.com/docs/@vigilante%2Fdocs-551-backtick-placeholder-syntax-in-generated-reference-docs-cli/reference/cli/server).

</details>

### Accuracy note — cases that do *not* drop on live

These render fine today, so they are **not** evidence of dropping (the
PR still wraps them for consistency / MDX-safety):

-
[`api/organizations`](https://coder.com/docs/reference/api/organizations)
and
[`api/provisioning`](https://coder.com/docs/reference/api/provisioning):
`{'tag1':'value1','tag2':'value2'}` renders verbatim — curly braces are
not an HTML tag.
- Tokens containing `/` or `|` are escaped by the renderer and stay
visible (as literal `<...>`): `<username/uuid/me>`, `<org-name/uuid>`,
`<owner/repo>`, `<draft|open|merged|closed>`,
`<created_by_me|shared_with_me>`. Backticks still improve their
readability, but they were never dropped.
 2026-07-14 13:39:58 -04:00
Paweł Banaszewski 3126306598 feat: add --aigateway-proxy-target flag (#27122) 
Adds `--aigateway-proxy-target` option to
`deploymentGroupAIGatewayProxy` that defines URL to which intercepted
requests should be forwarded to.
Forward URL used to be hardcoded to `coderAPI.AccessURL` pointing to
embedded Gateway. With addition of standalone AI Gateway this needs to
be configurable.

Renamed `aibridgeproxyd.Server.coderAccessURL` and `coderAccessPort` ->
`gatewayURL` and `gatewayPort` + option to better reflect reality.
 2026-07-14 16:40:22 +00:00
Michael Suchacz 55b06f14a3 feat: allow overriding advisor reasoning effort (#27196) 2026-07-14 11:01:51 +02:00
Marcin Tojek 162bd0e015 chore(docs): update release docs for v2.34.6 (#27215) 
Automated docs update for v2.34.6 release.

Created by `releasetui`.
 2026-07-14 10:50:29 +02:00
Marcin Tojek ec52dd274c chore(docs): update release docs for v2.35.2 (#27214) 
Automated docs update for v2.35.2 release.

Created by `releasetui`.
 2026-07-14 10:30:16 +02:00
Michael Suchacz 6f6d7539c8 feat: remove unused chat statuses pending, paused, and completed (#27064) 
The chatd state machine only recognizes `waiting`, `running`, `error`,
`requires_action`, and `interrupting`. Remove the unused `pending`,
`paused`, and `completed` values from the database enum, backend, SDK,
frontend, generated queries, and API docs.

Migration `000543_chat_status_remove_unused` remaps existing `pending`
rows to `running`, remaps `paused` and `completed` rows to `waiting`,
drops the obsolete `idx_chats_pending` index, and recreates
`chats_expanded` around the enum swap. It also removes the dead
`AcquireChats` query and all remaining query literals for the deleted
statuses.

**NOTE**: The enum swap can break chat queries from older replicas
during a mixed-version rollout because they still reference
`'pending'::chat_status`. Chats are experimental, so this PR accepts
that limited rollout window instead of adding a two-release expand and
contract sequence.

> This PR was authored by Mux (AI agent) on Mike's behalf.
 2026-07-13 20:28:28 +02:00
Callum Styan ad29777cb2 feat: NATS mTLS pubsub implementation (#26902) 2026-07-13 11:00:02 -07:00
Nick Vigilante 199b5936c1 fix(docs): replace invalid </br> tags and format swallowed placeholder URL (#27174) 
## Summary

Fixes two classes of invalid/broken HTML in hand-written docs. Both are
visible problems in today's rendered docs, independent of any
docs-engine work.

1. **`</br>` is not a real HTML tag.** `br` is a void element with no
closing form; browsers error-correct `</br>`, but it is invalid HTML.
Replaced all 15 usages with `<br />` across:
   - `docs/admin/templates/extending-templates/dynamic-parameters.md`
   - `docs/admin/users/idp-sync.md`
   - `docs/tutorials/best-practices/organizations.md`
2. **Browser-swallowed placeholder URL.** In
`docs/ai-coder/github-to-tasks.md`,
`https://<your-coder-url>/settings/external-auth` was unformatted, so
HTML renderers parse `<your-coder-url>` as an unknown tag and drop it.
The live docs currently render the broken text `re-authenticate at
https:///settings/external-auth`. Wrapped in backticks, matching every
other instance in the same file.

Table realignment noise in the diff is from `fmt/markdown` (`<br />` is
one character wider than `</br>`).

A repo-wide grep confirms no remaining `</br>` and no other unformatted
`https://<placeholder>` URLs in prose (other hits are inside code fences
or already backticked). The equivalent placeholder issues in
**generated** reference docs (CLI help strings, swagger annotations) are
intentionally out of scope and tracked separately in
[DOCS-551](https://linear.app/codercom/issue/DOCS-551/backtick-placeholder-syntax-in-generated-reference-docs-cli-help).

Tracking issue:
[DOCS-550](https://linear.app/codercom/issue/DOCS-550/fix-invalid-br-tags-and-browser-swallowed-placeholder-url-in-hand)

---

Created by Coder Agents on behalf of @nickvigilante.
 2026-07-13 10:41:38 -04:00
Thomas Kosiewski 9af3d31036 fix: relabel advisor max uses setting from per run to per turn (#27046) 2026-07-10 09:36:12 +02:00
Danielle Maywood d66e4d794f feat: add configurable reasoning effort to Coder agents (#26974) 2026-07-09 23:35:12 +01:00
Denis Afonso 20ee0068e6 docs(docs/install/releases): update release calendar for v2.35.1 (#27128) 
Update release calendar with the latest branch releases:

- v2.34.1 → v2.34.5 (Stable/ESR)
- v2.33.7 → v2.33.11 (Security Support)
- v2.32.6 → v2.32.10 (Not Supported)
- v2.29.16 → v2.29.19 (Extended Support Release)
- 2.35 added as Mainline at v2.35.1

Channel rotation for the 2.35 mainline release:
- 2.32: Security Support → Not Supported
- 2.33: Stable → Security Support
- 2.34: Mainline (ESR) → Stable (ESR)
- 2.35: Not Released → Mainline

Also updates the ESR version link to point to v2.34.5.
 2026-07-09 18:47:50 +01:00
Andrew Aquino 3e85cfb2c5 fix: during workspace bulk start/stop, skip workspaces already in target state (#27108) 
Previously, bulk start required every selected workspace to be stopped,
and bulk stop required every selected workspace to be running. Mixed
selections disabled both buttons entirely.

- Change the disabled checks on bulk start/stop from `every()` to
`some()` so the buttons are enabled when at least one workspace is
eligible.
- Filter workspaces by status in the mutation functions so only eligible
workspaces are sent to the API, matching the pattern used by other batch
mutations (update, favorite, unfavorite).
- Update docs to reflect the new behavior.

> [!NOTE]
> Generated by Coder Agents. [View session](https://coder.com/).

<details>
<summary>Implementation plan</summary>

## Problem

When an admin selects multiple workspaces and opens the "Bulk actions"
dropdown, the **Start** menu item is disabled unless *every* selected
workspace has `latest_build.status === "stopped"`. If even one workspace
is already running (or in any other non-stopped state), the Start button
is grayed out and unusable. Same issue applies to **Stop**.

## Changes

### 1. Relax disabled condition (`WorkspacesPageView.tsx`)

Changed `every()` to `some()` for both Start and Stop dropdown items.
The buttons are now enabled when at least one selected workspace is in
the target state.

### 2. Filter in mutations (`batchActions.ts`)

Added `.filter()` before `.map()` in both `startAllMutation` and
`stopAllMutation` so only eligible workspaces hit the API. This matches
the existing pattern in `updateAllMutation`, `favoriteAllMutation`, and
`unfavoriteAllMutation`.

### 3. Update documentation (`docs/user-guides/workspace-management.md`)

Replaced "can only be applied to a set of workspaces which are all in
the same state" with "apply to eligible workspaces in the selection,
skipping workspaces that are already in the target state."

## Testing

Four new Storybook stories:

| Story | What it tests |
|-------|---------------|
| `StartIgnoresAlreadyRunningWorkspaces` | Mixed selection; only stopped
workspaces get `startWorkspace` calls |
| `StopIgnoresAlreadyStoppedWorkspaces` | Mixed selection; only running
workspaces get `stopWorkspace` calls |
| `StartDisabledWhenNoWorkspacesAreStartable` | All running; Start
button is disabled |
| `StopDisabledWhenNoWorkspacesAreStoppable` | All stopped; Stop button
is disabled |

</details>
 2026-07-09 08:13:38 -07:00
Paweł Banaszewski bab8ce9d41 feat: setup logging, tracing and metrics in standalone AI Gateway (#27068) 
Adds logging, tracing and metrics setup to standalone AI Gateway.
Existing options are re-used when possible.
 2026-07-09 14:02:18 +00:00
Danny Kopping ef0b5585d5 feat: record and expose terminal upstream interception errors (#26961) 
Categorises the terminal error of a failed interception and persists it
on the interception record, then surfaces it on the AI Gateway API.

- Categorise into an enum (`bad_request`, `unauthorized`,
  `rate_limited`, `overloaded`, `server_error`, `unknown`), unwrapping
  the ResponseError envelope, the upstream Anthropic/OpenAI SDK errors,
  and key-pool exhaustion so blocking and streaming paths agree.
- Thread the type and raw message through the recorder dRPC into the
  `aibridge_interceptions` row (optional proto fields; NULL on success).
- Expose the error on the AI Gateway thread API from the root
  interception.

*This PR was produced by opencode (agent) using the `anthropic/claude-opus-4-8` model, under human direction and review.*
 2026-07-09 15:36:56 +02:00
Nick Vigilante 83cb587c3f docs(docs/.style/style-guide): add directional-language and contractions rules (#26729) 
Adds two new accessibility-and-voice rules to the style guide.

**Directional language**
(`docs/.style/style-guide/accessibility-and-inclusion.md`).
Screen-reader users navigate documents linearly and cannot follow
spatial references like "see below" or "the menu on the left". The rule
prescribes anchor links, section headings, document order ("the previous
section", "the following section"), and named UI elements instead. A
replacement table covers the common cases.

**Contractions are the default**
(`docs/.style/style-guide/voice-and-tone.md`). Prefer contractions in
body prose for the same reason the docs use second person and present
tense. Three exceptions: auxiliary contractions (`you'd`, `there's`,
`it's`, `we'd`, `they're`) need an explicit complement and cannot end a
sentence; contractions join exactly two words (no `you'd've` or
`wouldn't've`); spell out for emphasis and high-stakes operations like
deletion or data loss (`do not`, `cannot`, `will not`).

The PR also sweeps the existing style-guide subpages so the existing
prose comply with both rules.

Resolves
[DOCS-462](https://linear.app/codercom/issue/DOCS-462/add-screen-reader-aware-directional-language-rule-and-sweep-existing).

<details>
<summary>Directional-language sweep targets</summary>

| File | Change |
| --- | --- |
| `docs/.style/style-guide/README.md` | "pages below" becomes "linked
pages" |
| `docs/.style/style-guide/accessibility-and-inclusion.md` | "top of the
page" becomes "beginning of the page". Captions "follow" instead of "go
below". Latin abbreviation table cells drop "as described below". Sample
captions name widgets instead of panel positions. |
| `docs/.style/style-guide/audience-and-scope.md` | Don't example
rewritten without "below". "Above the first paragraph" becomes "before
the first paragraph". "At the top of the page" becomes "at the beginning
of the page". |
| `docs/.style/style-guide/capitalization-and-punctuation.md` |
"Exceptions above" becomes "exceptions listed earlier". |
| `docs/.style/style-guide/formatting.md` | Captions "follow" instead of
"go below". Sample captions renamed by widget. Don't example "as shown
above" becomes "as shown in the screenshot". |
| `docs/.style/style-guide/numbers-units-and-dates.md` | "10th and up"
becomes "10th and higher". |

Idiomatic stack metaphors like "built on top of Terraform" and phrasal
verbs like "set up", "back up", "log in", and "shut down" are explicitly
carved out as not directional and stay as-is.

</details>

<details>
<summary>Contractions rule scope</summary>

The rule lands as `## Contractions are the default` in
`voice-and-tone.md`, placed between `Present tense by default` and
`Trailing prepositions are a judgment call` because all three rules sit
in the natural-phrasing cluster.

The sweep applies the rule across all eight style-guide subpages: 76
lines updated where the spelled-out form (`does not`, `is not`,
`cannot`, `you have`, `there is`, `that is`) reads more naturally as a
contraction.

Skipped:

- Don't blocks inside the contractions rule that intentionally
demonstrate the wrong form.
- Do blocks inside the emphasis sub-rule that intentionally model `do
not`, `cannot`, and `will not` for high-stakes operations.
- The Churchill joke inside the trailing-prepositions Don't blocks.
- The "that is" dictionary definition of `i.e.` in the Latin
abbreviations table.
- `may not` (no contraction in modern English).
- `that has` relative clauses where `'s` could read as possessive.

</details>

<details>
<summary>Lints</summary>

- `make lint/markdown`: 0 errors across 495 files.
- `make lint/prose`: only the pre-existing intentional `[Demo]`
annotations in `docs/.style/_vale-annotation-demo.md` fire.

</details>

---

*Filed via [Coder Agents](https://coder.com/docs/ai-coder/agents) on
Nick's behalf.*
 2026-07-08 12:05:24 -04:00
Nick Vigilante f7632451f4 feat(docs): add Coder.BrandNames Vale rule, enforce HashiCorp casing (#25501) 
Lands the first concrete rule under the `Coder` style:
`Coder.BrandNames`, a bundled `substitution` rule that enforces
canonical brand casing in prose. HashiCorp is the first entry;
[DOCS-188](https://linear.app/codercom/issue/DOCS-188) extends it with
GitHub, OpenTofu, Kubernetes, Terraform, JetBrains, and VS Code.

## What changes

Four commits, ordered so each is independently valid:

1. **`docs: fix HashiCorp casing in prose and sidebar`**
([06d769dad1](https://github.com/coder/coder/pull/25501/commits/06d769dad179cf85c535b058df3b6bafdc1f9565)).
5 Markdown files plus 2 `docs/manifest.json` entries. Drives the corpus
violation count to zero.
2. **`feat(docs/.style/styles/Coder): add Coder.BrandNames Vale rule`**
([e00fc780a7](https://github.com/coder/coder/pull/25501/commits/e00fc780a7a20dcf82105d997af5cfcddd4b1855)).
New `BrandNames.yml` with the HashiCorp swap at `level: error`, plus a
new `### Brand names` subsection in `docs/.style/style-guide.md`.
3. **`docs(.style/styles/Coder/README.md): scrub planned-rules notes
obsoleted by Coder.BrandNames`**
([af8833b9f5](https://github.com/coder/coder/pull/25501/commits/af8833b9f58dd617732ae533bbf53eb4fc2e816a)).
Removes the README's "intentionally empty for now" lead-in and the
obsolete HashiCorp casing bullet from the planned-coverage list.
4. **`docs: apply semantic line breaks and fix Vale findings on
PR-touched files`**
([e9f11df188](https://github.com/coder/coder/pull/25501/commits/e9f11df1886fdf0d5efc5e8a2cab95fecbd898f9)).
Pre-review pass on every Markdown file this PR modifies. Full sembr and
Vale-warning cleanup on the style-guide infrastructure
(`style-guide.md`, `Coder/README.md`); sembr applied to the HashiCorp
swap paragraph only on the five product docs, per scoping discussion
with @nickvigilante.

## Severity rationale

`error` from day one. HashiCorp's brand owner publishes a canonical
casing; any other casing in prose is wrong, not a judgment call. Matches
the `error = low FPs x high gravity` framework. False-positive rate is
effectively zero because Vale's `substitution` rule skips inline code,
fenced code blocks, and URLs by default, so `hashicorp/kubernetes`
(Terraform provider source) and `developer.hashicorp.com` stay
untouched.

## Verification

- `make lint/markdown`: 0 errors across 487 files.
- `make lint/prose`: 1 error, 1 warning, 1 suggestion in 468 files. All
three findings are the intentional `Coder.DemoError`,
`Coder.DemoWarning`, and `Coder.DemoSuggestion` annotations on
`docs/.style/style-guide/demo/demo.md` (added on main as part of the
[DOCS-425](https://linear.app/codercom/issue/DOCS-425) inline-annotation
demo), not real findings. `Coder.BrandNames` fires zero times against
the cleaned-up corpus.
- `make pre-commit-light`: passed (7s).
- Self-test: ran the rule against an unmodified `docs/` and confirmed it
flags the 7 prose instances the cleanup commit fixes, then re-ran
against the post-cleanup state and confirmed zero alerts.

## Known future conflict

When [#26632](https://github.com/coder/coder/pull/26632)
([DOCS-434](https://linear.app/codercom/issue/DOCS-434)) merges, the
monolithic `docs/.style/style-guide.md` is split into the
`docs/.style/style-guide/` multi-page structure. The `### Brand names`
subsection added in commit 2 will need to land in
`docs/.style/style-guide/word-choice.md` (which already references the
rule), and the `link:` in `docs/.style/styles/Coder/BrandNames.yml` will
need to update from `style-guide.md#brand-names` to
`style-guide/word-choice.md#brand-names`. Resolution path documented in
an inline comment on this PR.

<details>
<summary>Implementation plan and decision log</summary>

### Why bundle into Coder.BrandNames rather than one file per brand

Vale's convention (mirrored by `Google.WordList` with ~70 swaps in a
single file) is to bundle `substitution` rules when they share severity,
message template, and link. All brand-name rules share that shape:
`error`, `Use '%s' instead of '%s'`, link to the style guide section.
Bundling reduces "add a brand" to a one-line YAML diff and keeps
`CODEOWNERS` and blame coherent. Per-rule performance is irrelevant at
this scale; Vale's per-rule overhead is sub-millisecond and dwarfed by
Markdown parsing.

### Why the cleanup lands first

Commits are ordered cleanup-then-rule so each commit is a known-good
state:

- After commit 1: corpus is HashiCorp-clean, but no rule exists yet.
- After commit 2: rule exists and lints a clean corpus.

Reversing the order would land the rule at commit 1 (firing 7 errors on
uncleaned content) and resolve them at commit 2. Under `--no-exit` the
CI job still passes, but the inline annotations on commit 1 would be
misleading.

### Why HashiCorp first instead of all brands at once

Proof-of-concept value. HashiCorp is the smallest cleanup (7 prose lines
plus 2 sidebar lines = 9 lines), zero FPs, zero ambiguity. Once the loop
(rule plus cleanup plus style-guide section) is proven,
[DOCS-188](https://linear.app/codercom/issue/DOCS-188) appends the other
brands as additional commits to the same bundle.

### Brand-token sensitivity

The `swap:` table only matches:

- `Hashicorp` (capital H, lowercase rest), the actual wrong form in the
corpus.
- `HASHICORP` (all caps), defensive; doesn't appear in current corpus
but cheap to include.

`hashicorp` (all lowercase) is **not** in the swap table. The lowercase
form appears 49 times in URLs (`developer.hashicorp.com`,
`registry.terraform.io/providers/hashicorp/...`,
`github.com/hashicorp/...`) and 6 times as Terraform provider sources
(`source = "hashicorp/kubernetes"`), all of which are correct lowercase
by convention. Vale's substitution rule scope ensures URLs and code
blocks are skipped, but skipping the rule entirely for `hashicorp`
(lowercase) is the explicit decision; if a prose typo of lowercase
"hashicorp" ever shows up, we'd catch it through `Vale.Spelling`
([DOCS-187](https://linear.app/codercom/issue/DOCS-187)) instead.

### Self-reference in the style guide

The `### Brand names` section's example table needed `Hashicorp` and
`HashiCorp` as literal demonstration tokens. Wrapping them in backticks
(`` `Hashicorp` ``, `` `HashiCorp` ``) keeps Vale from flagging the
wrong-case example as a real violation. This is correct typography too:
demonstration tokens get code formatting.

### Manifest.json

Vale doesn't lint JSON, so the two `docs/manifest.json` entries are
fixed by direct edit rather than tool enforcement. The sidebar `path`
(`./admin/integrations/vault.md`) is unchanged; the title change does
not affect the page URL on coder.com. No redirect needed in
`coder/coder.com:redirects.json`.

### Pre-mortem

- **Generated docs noise**: `Coder.BrandNames` does not fire on
auto-generated `docs/reference/` content because no codersdk identifier
matches the swap pattern. Zero risk.
- **Future-additions friction**: adding GitHub to the swap table is one
YAML line and a cleanup commit. The bundling shape pays off here.
- **Disable footgun**: if a contributor needs to write the wrong casing
on purpose (quoting an external bug report verbatim, for example), they
can wrap the literal in backticks (already correct typography) or use
the per-line Vale skip comment.

</details>

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

---

*Filed via [Coder Agents](https://coder.com/docs/ai-coder/agents) on
Nick's behalf.*
 2026-07-08 11:43:29 -04:00
Nick Vigilante 83acdaebd1 docs: add DOCKER_HOST guidance for non-default Docker socket paths (#26807) 
## What

Add `DOCKER_HOST` guidance for non-default Docker socket paths to two
pages:

- `docs/install/docker.md`: expands the **Cannot connect to the Docker
daemon**
troubleshooting section with the `DOCKER_HOST` fix and how to persist it
to
  your shell startup file.
- `docs/admin/templates/troubleshooting.md`: adds a concise **Cannot
connect to
the Docker daemon** entry that cross-references the install guide for
the full
  steps.

## Why

`install/docker.md` previously documented only the default socket path
(`/var/run/docker.sock`). When Docker runs through a tool that uses a
per-user
socket, such as rootless Docker on Linux, or Colima, Podman, or Rancher
Desktop
on macOS, the daemon exposes its socket at a non-default path, so the
Coder
server cannot connect until `DOCKER_HOST` is set. The guidance frames
Colima as
one example, notes that default socket paths vary by tool, and persists
the
setting in a shell-agnostic way.

Generated by Coder Agents on behalf of @nickvigilante.
 2026-07-08 11:33:02 -04:00
Susana Ferreira 48f07e6e13 feat: add user AI spend endpoint (#26978) 
## Description

Adds the `GET /api/v2/users/{user}/ai/spend` endpoint returning the
user's current AI spend, effective budget, and period bounds.

## Changes

- Add `userAISpendStatus` handler under the same feature/experiment gate
as `/api/v2/users/{user}/ai/budget`.
- Add `codersdk.UserAIBudgetSummary` (embedded into `UserAISpendStatus`)
and a `UserAISpendStatus` client method.
- Move `LimitSource` from `coderd/aibridge/budget` to `codersdk` so the
type is shared across endpoints.

Closes https://linear.app/codercom/issue/AIGOV-472

> [!NOTE]
> Initially generated by Claude Opus 4.7, modified and reviewed by
@ssncferreira
 2026-07-08 12:46:47 +01:00
Paweł BanaszewskiandDanny Kopping ccba3969ab feat: add ai-gateway start command (#26605) 
> AI Tools were used to produce this PR

This PR adds `coder ai-gateway start` command that runs the AI Gateway
as an independent process.

- Standalone process doesn't have access to DB. Uses DRPC services under
`/api/v2/ai-gateway/serve`for auth, recording and provider
initialization.
- It only handles LLM traffic, other endpoints (eg. `/sessions`) are
only available though `coderd`.
- The standalone gateway reuses applicable flags from AI Gateway
deployment options. Provider-seeding and coderd-only options are
excluded.
- Only added to fat build, the slim build stub rejects the command.

Some wiring used by this new command is added.

**`NewWebsocketDialer`** - implements the standalone gateway's
connection to coderd's `/api/v2/ai-gateway/serve` endpoint. It upgrades
to a WebSocket, multiplexes with yamux, and wires all DRPC services.

**`AIGatewayDataPlaneMiddleware`** - extracts the per-request middleware
chain (concurrency limiting, rate limiting, BYOK gating) into a shared
function used by both the embedded route and the standalone gateway.

**`RootCmd.ResolveClientConnection`** - resolve the deployment URL and
builds an HTTP transport without requiring a session token. Used in
`ai-gateway start`command as it authenticates using different credential
type.

---------

Co-authored-by: Danny Kopping <danny@coder.com>
 2026-07-08 11:12:53 +02:00
Cian Johnston 14767dfa8e feat: explain missing POSIX sh on Windows workspaces (#27048) 
When a workspace has no POSIX sh on PATH (typical for fresh Windows workspaces), the execute tool fails with a raw `exec: "sh": executable file not found in %PATH%` error the model cannot act on. 

This change:
- Enriches the above error in chattool with remediation steps and a docs link.
- Documents the requirement in the Coder Agents architecture page.

> This PR was generated by Coder Agents on behalf of @johnstcn
 2026-07-07 21:32:27 +00:00
Bobby Ho b169f4d8cb feat: expose external auth token expiry in agent API and CLI (#26883) 
Previously, \`ExternalAuthResponse\` contained no expiry information, so
workspace agents and git credential helpers had no way to know when a
cached token would stop being valid. Every git operation had to call
back to coderd via \`GIT_ASKPASS\` to get a fresh token, adding 1-2
seconds of latency.

This PR surfaces \`OAuthExpiry\` from the database as \`ExpiresAt\` in
\`ExternalAuthResponse\`, allowing agents to cache tokens with correct
eviction timing (compatible with \`git-credential-cache --timeout\` and
\`password_expiry_utc\` introduced in git 2.34).

\`ExpiresAt\` is normalized to UTC before JSON encoding to avoid
sub-minute precision loss that occurs when the PostgreSQL driver applies
historical Local Mean Time (LMT) timezone offsets to year-1 AD
timestamps.

The \`coder external-auth access-token\` CLI command gains \`--output
json\` to print the full response including \`ExpiresAt\`, enabling
scripts to consume the expiry without parsing heuristics.

Closes https://github.com/coder/coder/issues/26036

## Manual Test

<details>
<summary>Setup</summary>

1. Create a GitHub OAuth app at https://github.com/settings/developers
with:
   - Homepage URL: `http://127.0.0.1:3000`
- Authorization callback URL:
`http://127.0.0.1:3000/external-auth/github/callback`

2. Start the dev server with the GitHub provider configured:
   ```sh
CODER_EXTERNAL_AUTH_0_ID=github CODER_EXTERNAL_AUTH_0_TYPE=github
CODER_EXTERNAL_AUTH_0_CLIENT_ID=<client-id>
CODER_EXTERNAL_AUTH_0_CLIENT_SECRET=<client-secret> ./scripts/develop.sh
   ```

3. Log in at `http://127.0.0.1:3000` (use `127.0.0.1`, not `localhost`,
so the OAuth state cookie domain matches the callback URL).

4. Go to Account > External Authentication and click **Connect** next to
GitHub. Complete the OAuth flow.

5. Create a workspace and SSH into it:
   ```sh
   coder create test-workspace
   coder ssh test-workspace
   ```

</details>

<details>
<summary>Flow 1: Token is valid — JSON output includes
<code>expires_at</code></summary>

Inside the workspace, run:

```sh
coder external-auth access-token github --output json
echo "Exit code: $?"
```

Expected output (GitHub tokens have no expiry, so \`expires_at\` is the
zero value):

```json
{
  "access_token": "<redacted>",
  "token_extra": null,
  "url": "",
  "type": "github",
  "expires_at": "0001-01-01T00:00:00Z",
  "username": "<redacted>",
  "password": ""
}
```

```
Exit code: 0
```

</details>

<details>
<summary>Flow 2: Token missing — JSON output includes auth URL, exit
code 1</summary>

Disconnect GitHub in the Coder UI (Account > External Authentication >
Disconnect), then inside the workspace run:

```sh
coder external-auth access-token github --output json
echo "Exit code: $?"
```

Expected output:

```json
{
  "access_token": "",
  "token_extra": null,
  "url": "http://127.0.0.1:3000/external-auth/github",
  "type": "",
  "expires_at": "0001-01-01T00:00:00Z",
  "username": "",
  "password": ""
}
```

```
Exit code: 1
```

</details>
 2026-07-07 12:38:37 -07:00
George K 6af0f4d698 feat: add workspace restart functionality to API (#25757) 
This models restart as durable orchestration of existing stop and
start workspace builds instead of adding a new restart transition.
Keeping restart as two existing transitions preserves the current
build/provisioner model.

The child start build is created only after the parent stop build
succeeds, rather than being inserted immediately in a pending
state. That keeps `workspace_builds` aligned with actual
provisioner-ready work and avoids introducing a second
pending-build lifecycle that the provisioner and build acquisition
paths would need to understand.

Refs: https://linear.app/codercom/issue/PLAT-143
 2026-07-07 09:18:30 -07:00
Sas Swart fc188fdaee fix: create agent firewall sessions without requiring agent read access (#26990) 
## Overview

Part of the **boundary correlation** feature. Fixes lazy creation of
`boundary_sessions` rows so it works within the agent's RBAC
constraints, and consumes the new `ConfinedProcessName` field reported
by boundary.

Pairs with coder/boundary#206, which adds `ConfinedProcessName` to
`ReportBoundaryLogsRequest`. This branch bumps the
`github.com/coder/boundary` module to pick up that work.

## Problem

`ensureSession` did a pre-insert existence check via
`GetBoundarySessionByID`. Agents are **not permitted to read boundary
sessions**, so that read path is not viable when the session is created
from an agent-reported log batch.

## Changes

- **Remove the pre-insert read.** `ensureSession` now inserts directly
and treats a primary-key unique violation as success, covering sessions
already created by a prior batch, a reconnection, or another coderd
replica — without requiring read access.
- **Per-connection guard.** Add a mutex-protected `ensuredSessions` set
so repeated log batches on the same connection skip the existence check
and insert entirely, touching the database only for the logs. On a
transient insert failure the session is left unmarked so the next batch
retries.
- **Consume `ConfinedProcessName`.** Pass `req.GetConfinedProcessName()`
through to the session insert.
- **Bump boundary module** from `v0.9.0` to
`v0.9.1-0.20260706095856-35ba90f9e8b2`.
- **Tests.**
- Add `TestReportBoundaryLogsAgentRBAC`
(`coderd/boundary_logs_test.go`), an integration test that connects as a
real workspace agent, verifies the session and log are persisted under
agent RBAC, and asserts the agent subject cannot read boundary sessions
— guarding against reintroducing a pre-insert read.
- Add `TestReportBoundaryLogsSessionGuard` (session inserted once across
two batches, logs inserted per batch) and
`TestReportBoundaryLogsSessionRetriedOnError` (insert retried after a
transient error).
- Regenerate `agent-firewall` CLI docs/golden files and adjust the
clidocgen template to render the YAML path when a flag has no long name.

> 🤖 This PR was opened by Coder Agents on behalf of @SasSwart.
 2026-07-07 10:42:01 +00:00
Danielle Maywood d51762440b feat: add custom AI provider icons and instance-based model picker grouping (#27026) 2026-07-06 23:00:09 +01:00
Ben PotterandJeremy Ruppel 7b19ec3933 feat: improve the image management experience with template builder (#27018) 
Makes it easier to pick the right workspace image, both in the template
builder and in the docs.

- Template builder: the Docker and Kubernetes bases now expose a
`container_image` variable in the wizard (freeform text, defaults to
`codercom/example-base:ubuntu`), and their prerequisites explain why
image choice matters, with tradeoffs between
`codercom/example-base:ubuntu` (minimal) and
`codercom/example-universal:ubuntu` (catch-all), plus pointers to
[coder/images](https://github.com/coder/images) and the image management
docs.
- Docs: reworked [image
management](https://coder.com/docs/@ben%2Fdevrel-201-image-guidance-prereqs/admin/templates/managing-templates/image-management)
into a clearer maturity ladder (minimal → golden → project-specific →
developer customization), with pullable image references in every
example, `codercom/oss-dogfood` as a project-specific example, and Dev
Containers + [mise](https://mise.jdx.dev/) as ways to customize without
new images.

Companion PR for the starter templates: coder/registry#943

Part of DEVREL-201.

🤖 Generated with Coder Agents using Claude, on behalf of @bpmct (wizard
variable by @jeremyruppel in #27024)

---------

Co-authored-by: Jeremy Ruppel <jeremyruppel@users.noreply.github.com>
 2026-07-06 20:37:34 +00:00
Ironbelly 2de8b3c74a Merge pull request #25638 from ironbelly/docs/cli-ssh-argv-quoting 
docs: clarify argv quoting behavior with `--`
 2026-07-06 16:32:03 +00:00
Jeremy Ruppel 79fc8541ed docs: update template creation docs for template builder (#26993) 
## Summary

Update documentation across 9 files to present the template builder as
the primary template creation method, replacing the old starter
templates flow as the default entry point.

The template builder is a guided wizard that lets admins select base
infrastructure, add registry modules, configure variables, and produce
validated Terraform without writing HCL.

## Changes

**Primary docs (significant rewrites):**
- `docs/admin/templates/creating-templates.md`: Added "Using the
template builder" as the first section with full 5-step wizard
documentation, screenshots, airgap/registry notes, and alternative
creation links. Moved CLI starter template flow to its own section.
Fixed "You can the" typo.
- `docs/get-started/index.md`: Rewrote Steps 4-6 to use the builder with
the Docker base template instead of the Coder Quickstart (which is not a
builder base template). Generalized workspace parameter instructions.
- `docs/start/first-template.md`: Rewrote to use the builder. Removed
old starter templates references, TODO notes, typo, and commented-out
sections.

**Secondary docs (targeted edits):**
- `docs/admin/templates/index.md`: Replaced starter templates section
with builder-first "Create a template" section.
- `docs/admin/templates/managing-templates/index.md`: Renamed "Starter
templates" to "Creating templates" pointing to the builder.
- `docs/install/airgap.md`: Added "Template builder" section documenting
`CODER_DISABLE_TEMPLATE_BUILDER` and
`CODER_TEMPLATE_BUILDER_REGISTRY_URL`.
- `docs/tutorials/template-from-scratch.md`: Added TIP callout
recommending the builder. Fixed `coder templates create` -> `coder
templates push` inconsistency.
- `docs/admin/integrations/devcontainers/envbuilder/add-envbuilder.md`:
Updated Dashboard tab to reference the builder and "Upload an existing
template" alternative.
- `docs/about/screenshots.md`: Updated caption and image reference for
template builder.

**Screenshots added:**
- `templatebuilder_01_bases.png` (base selection step)
- `templatebuilder_02_modules.png` (module selection step)
- `templatebuilder_03_module_customization.png` (module settings step)
- `templatebuilder_04_customizations.png` (template customizations step)

<details>
<summary>Implementation plan</summary>

# Plan: Update docs/ for Template Builder Launch

## Summary

The Template Builder is a new guided wizard at `/templates/new/builder`
that lets admins create templates by selecting a base infrastructure
template, composing it with registry modules, configuring variables, and
producing a validated Terraform bundle without writing HCL. The docs
need to be updated to present this as the primary/recommended template
creation path, while preserving the existing paths (upload, CLI,
duplicate) as alternatives.

## Key behavioral facts from the code

- **Route**: `/templates/new/builder` (new), `/templates/new` (old,
still exists)
- **Entry point**: The "New Template" button on the Templates page links
to `/templates/new/builder` when the builder is enabled; otherwise falls
back to `/starter-templates`
- **5-step wizard**:
  1. **Select base infrastructure** (e.g., Docker, AWS EC2, Kubernetes)
  2. **Base template parameters** (optional, skipped if base has none)
3. **Select modules** (IDE, AI Agent, Source Control, etc.;
multi-select, grouped by category)
4. **Module settings** (optional, skipped if no configurable variables)
5. **Template customizations** (name, display name, description, icon,
organization)
- **Alternative creation links** are shown on step 1: "Start from
scratch", "Upload an existing template", "Browse community templates",
"Use template agent skill"
- **Disabled via**: `CODER_DISABLE_TEMPLATE_BUILDER` env var /
`--disable-template-builder` flag. When disabled, redirects to old
`/templates/new` flow
- **Registry URL override**: `CODER_TEMPLATE_BUILDER_REGISTRY_URL`
(default: `registry.coder.com`)
- **Requires outbound access** to `registry.coder.com` for `terraform
init` at compose time
- **Modules are bundled** with the Coder release binary; the builder
does not fetch metadata from the registry at runtime
- **Sensitive variables** (secrets) are not collected by the builder;
they are deferred to workspace creation time
- **Module conflicts** show a warning but do not block creation
- **One-way**: No re-entry into the builder for existing templates; edit
HCL directly after creation

## Files to update

### Tier 1: Primary creation flow docs (significant rewrites)

#### 1. `docs/admin/templates/creating-templates.md`

**Current state**: Documents three creation paths: "From a starter
template" (primary), "From an existing template", "From scratch
(advanced)".

**Changes**:
- Add a new section **"Using the template builder"** as the first and
primary section (before "From a starter template").
- Describe the 5-step wizard flow: select base infrastructure, configure
base parameters, select modules, configure module settings, set template
customizations.
- Mention that the builder is enabled by default and requires outbound
access to `registry.coder.com`.
- Note that sensitive variables are collected from developers at
workspace creation, not during template building.
- Add a callout about disabling the builder for airgapped deployments
(`CODER_DISABLE_TEMPLATE_BUILDER`).
- Note the `CODER_TEMPLATE_BUILDER_REGISTRY_URL` option for self-hosted
registry mirrors.
- Keep existing "From a starter template", "From an existing template",
and "From scratch" sections largely intact, but reframe them as
alternative paths.
- Update the "From a starter template" Web UI instructions to note the
new entry point routing (the "New Template" button now goes to the
builder when enabled).
- Fix existing typo: "You can the [Coder CLI]" should be "You can use
the [Coder CLI]".

#### 2. `docs/start/first-template.md`

**Current state**: Beginner tutorial walking through creating a template
from the Docker starter template via the old flow. Has a typo (`s` at
end of line 32), commented-out workspace creation section, and TODO
notes.

**Changes**:
- Rewrite steps 2 and 3 to use the Template Builder as the primary path.
- Step 2: Navigate to **Templates**, select **New Template**, which
opens the Template Builder.
- Step 3: Walk through the builder wizard steps (select Docker base,
optionally select modules like code-server, configure template
name/description, create).
- Remove the typo on line 32 (`s`).
- Keep the "Modify your template" section (step 6) intact since it
covers post-creation editing which is unchanged.
- Remove or update the reference to "Starter Templates" as a separate
page since the builder subsumes that entry point.

#### 3. `docs/get-started/index.md`

**Current state**: Quickstart guide. Step 4 says "Select **Templates** →
**New Template**" then pick "Coder Quickstart" from starter templates.

**Changes**:
- Update Step 4 to describe using the Template Builder.
- The flow becomes: Select **Templates** → **New Template** → builder
opens → select **Coder Quickstart** as the base template → optionally
add modules → set name/description → **Create Template**.
- Update the "What just happened?" explanation to mention the builder
composed and validated the Terraform.
- Screenshot reference `create-quickstart-template.png` will need a new
screenshot (note this in the PR; screenshots are out of scope for this
change but should be flagged).

### Tier 2: Secondary references (targeted edits)

#### 4. `docs/admin/templates/index.md`

**Current state**: Overview page mentioning starter templates as the
primary creation path.

**Changes**:
- Update the "Starter templates" section to mention the Template Builder
as the recommended way to create templates, with starter templates
serving as base templates within the builder.
- Update the link to point to the builder section: `[Create a template
with the template
builder](./creating-templates.md#using-the-template-builder)`.
- Update the screenshot reference and caption. The "Starter Templates"
page screenshot may no longer be the first thing admins see.

#### 5. `docs/admin/templates/managing-templates/index.md`

**Current state**: Documents starter templates, editing, updating,
deleting.

**Changes**:
- Update the "Starter templates" section to mention the Template Builder
as the primary creation path, with starter templates available as base
templates within it.
- Update the image reference from `starter-templates.png` if it shows
the old flow.

#### 6. `docs/tutorials/template-from-scratch.md`

**Current state**: Detailed tutorial for writing a template from scratch
with Terraform.

**Changes**:
- Add a brief note at the top recommending the Template Builder for
users who want to create templates without writing Terraform, with a
link to
`docs/admin/templates/creating-templates.md#using-the-template-builder`.
- In section "7. Create the template in Coder" → "Dashboard" tab, update
the UI steps. The "Upload template" option is now accessed via the old
creation flow at `/templates/new` (or through the "Upload an existing
template" link in the builder's alternatives).
- Fix the inconsistency where text says `coder templates create` but the
code block uses `coder templates push`.

#### 7.
`docs/admin/integrations/devcontainers/envbuilder/add-envbuilder.md`

**Current state**: Documents creating envbuilder templates via
Dashboard, CLI, and Registry tabs.

**Changes**:
- In the Dashboard tab, update the instructions. The "Create Template"
button now opens the builder by default. Users need to use the "Upload
an existing template" alternative link or navigate to `/templates/new`
directly.
- Update "From scratch" reference since that option is now an
alternative link in the builder.
- The CLI and Registry tabs remain unchanged.

#### 8. `docs/install/airgap.md`

**Current state**: Documents air-gapped installations. No mention of
Template Builder.

**Changes**:
- Add a note in the relevant section about the Template Builder
requiring outbound access to `registry.coder.com`.
- Document `CODER_DISABLE_TEMPLATE_BUILDER` for fully air-gapped
deployments.
- Document `CODER_TEMPLATE_BUILDER_REGISTRY_URL` for deployments using a
self-hosted registry mirror.

#### 9. `docs/about/screenshots.md`

**Current state**: Contains a caption "Template administrators can
either create a new Template from scratch or choose a Starter Template".

**Changes**:
- Update the caption to mention the Template Builder as the primary
creation method.
- Screenshot reference may need updating (flag for new screenshot).

### Tier 3: Minor/link-only updates

#### 10. `docs/admin/users/organizations.md`

- If it references the old "Create Template" screen with an org picker,
add a note that the Template Builder also includes organization
selection in its final step.

#### 11. `docs/ai-coder/tasks.md`

- If it mentions creating templates, add a passing reference to the
Template Builder as an option.

## Files NOT to update

- `docs/reference/api/templatebuilder.md`: Auto-generated API reference.
Already correct.
- `docs/reference/api/schemas.md`: Auto-generated. Already correct.
- `docs/reference/cli/server.md`: Auto-generated. Already has
`--disable-template-builder` and `--template-builder-registry-url`.
- `docs/reference/cli/templates_create.md`: Already deprecated.
- `docs/reference/cli/templates.md`: No changes needed.

## Implementation order

1. `docs/admin/templates/creating-templates.md` (primary creation docs,
most content)
2. `docs/get-started/index.md` (quickstart)
3. `docs/start/first-template.md` (beginner tutorial)
4. `docs/admin/templates/index.md` (overview)
5. `docs/admin/templates/managing-templates/index.md` (managing
overview)
6. `docs/install/airgap.md` (airgap note)
7. `docs/tutorials/template-from-scratch.md` (from-scratch tutorial)
8. `docs/admin/integrations/devcontainers/envbuilder/add-envbuilder.md`
(envbuilder)
9. `docs/about/screenshots.md` (screenshot captions)
10. Minor link/reference updates in tier 3 files

## Style notes

- Follow the Diataxis framework; keep tutorials as tutorials, reference
as reference.
- Use present tense, active voice, second person.
- Bold for UI elements: **Templates**, **New Template**, **Create
Template**.
- No emdash/endash.
- Do not add screenshots; flag where new screenshots are needed as
comments/TODOs.
- Run `make fmt/markdown` and `make lint/markdown` after all changes.
- Verify all pages are already in `docs/manifest.json` (no new pages
being added, only existing pages being updated).

</details>

> 🤖 Generated by Coder Agents
 2026-07-06 11:15:59 -04:00
Danielle Maywood 7c78698a6d feat: notify users when chats are shared (#26914) 2026-07-06 13:15:58 +01:00
Cian Johnston b21e0717d5 feat: remove chat chain mode (#26980) 
Removes OpenAI Responses "chain mode" from chatd. Closes CODAGT-445.

- Deletes `chatopenai/responses.go` (chain detection, activation, prompt filtering, response ID extraction) and its tests.
- Deletes the `ChainBroken` classification in `chaterror` and the chatloop retry bookkeeping that disabled chain mode mid-generation.
- Drops the `chain_broken` label from the `coderd_chatd_stream_retries_total` metric.
- Stops reading and writing `chat_messages.provider_response_id`
- Deletes the dead `ClearChatMessageProviderResponseIDsByChatID` query. Dropping the column is a follow-up migration.
- Deletes three chatloop hooks no caller sets (`ReloadMessages`, `DisableChainMode`, `PrepareMessages`), the dead `const AgentChatContextSentinelPath`, and stale chain-mode comments.

🤖 Generated by Coder Agents on behalf of @johnstcn.
 2026-07-06 11:57:12 +01:00
Itay Dafna d7ad85f7f6 feat: support multiple OIDC redirect URIs (#25408) 
This PR adds a new opt-in setting, `CODER_OIDC_REDIRECT_ALLOWED_HOSTS`,
that lets a single Coder deployment complete OIDC login on more than one
hostname. When the allowlist is non-empty, Coder picks the OIDC
`redirect_uri` based on the incoming request's Host header (validated
against the list) instead of always using the static URL derived from
`CODER_ACCESS_URL`. When unset, the (default) behavior is identical to
today.

The motivation is that a single Coder deployment is frequently reachable
via multiple hostnames - for example, an internal hostname for users on
a corporate VPN and a different hostname routed through a zero-trust
gateway for users off-VPN - but OIDC login today only works on whichever
single hostname `CODER_ACCESS_URL` points to, because the `redirect_uri`
sent to the IdP is fixed at server startup. Users who reach the
deployment on any other valid hostname can see the login page but fail
the OIDC callback, since the IdP redirects them back to a hostname they
can't reach (or whose cookies they don't have).
 2026-07-05 06:36:33 +02:00