Skip to content

Commit dfb3215

Browse files
[docs] Update documentation for features from 2026-03-01 (#18982)
1 parent f8dddc5 commit dfb3215

File tree

2 files changed

+39
-2
lines changed

2 files changed

+39
-2
lines changed

docs/src/content/docs/reference/github-tools.md

Lines changed: 4 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -57,6 +57,10 @@ tools:
5757

5858
## Guard Policies
5959

60+
:::caution[Experimental Feature]
61+
Guard policy fields (`repos` and `min-integrity`) are experimental and may change in future releases. Using either field emits a compilation warning.
62+
:::
63+
6064
Restrict which repositories and integrity levels the GitHub MCP server can access during agent execution. Guard policies apply fine-grained access control at the MCP gateway level.
6165

6266
```yaml wrap

docs/src/content/docs/reference/safe-outputs.md

Lines changed: 35 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -796,14 +796,20 @@ safe-outputs:
796796

797797
Resolves review threads on pull requests. Allows AI agents to mark review conversations as resolved after addressing the feedback. Uses the GitHub GraphQL API with the `resolveReviewThread` mutation.
798798

799-
Resolution is scoped to the triggering PR only — the handler validates that each thread belongs to the triggering pull request before resolving it.
799+
By default, resolution is scoped to the triggering PR. Use `target`, `target-repo`, and `allowed-repos` for cross-repository thread resolution.
800800

801801
```yaml wrap
802802
safe-outputs:
803803
resolve-pull-request-review-thread:
804-
max: 10 # max threads to resolve (default: 10)
804+
max: 10 # max threads to resolve (default: 10)
805+
target: "triggering" # "triggering" (default), "*", or number
806+
target-repo: "owner/repo" # cross-repository
807+
allowed-repos: ["org/repo1", "org/repo2"] # additional allowed repositories
808+
github-token: ${{ secrets.SOME_CUSTOM_TOKEN }} # optional custom token for permissions
805809
```
806810

811+
See [Cross-Repository Operations](/gh-aw/reference/cross-repository/) for documentation on `target-repo`, `allowed-repos`, and cross-repository authentication.
812+
807813
**Agent output format:**
808814

809815
```json
@@ -1219,6 +1225,33 @@ See [Cross-Repository Operations](/gh-aw/reference/cross-repository/) technical
12191225

12201226
## Global Configuration Options
12211227

1228+
### Workflow Call Outputs (`workflow_call`)
1229+
1230+
When a workflow uses `on: workflow_call` (or includes `workflow_call` in its triggers) and configures safe outputs, the compiler automatically injects `on.workflow_call.outputs` exposing the results of each configured safe output type. This makes gh-aw workflows composable building blocks in larger automation pipelines.
1231+
1232+
The following named outputs are exposed for each configured safe output type:
1233+
1234+
| Safe Output Type | Output Names |
1235+
|---|---|
1236+
| `create-issue` | `created_issue_number`, `created_issue_url` |
1237+
| `create-pull-request` | `created_pr_number`, `created_pr_url` |
1238+
| `add-comment` | `comment_id`, `comment_url` |
1239+
| `push-to-pull-request-branch` | `push_commit_sha`, `push_commit_url` |
1240+
1241+
These outputs are automatically available to calling workflows without any additional frontmatter configuration. User-declared `outputs` in the frontmatter are preserved and take precedence over the auto-injected values.
1242+
1243+
**Example — calling workflow using safe-output results:**
1244+
1245+
```yaml wrap
1246+
jobs:
1247+
run-agent:
1248+
uses: ./.github/workflows/my-agent.lock.yml
1249+
follow-up:
1250+
needs: run-agent
1251+
steps:
1252+
- run: echo "Created issue ${{ needs.run-agent.outputs.created_issue_number }}"
1253+
```
1254+
12221255
### Group Reports (`group-reports:`)
12231256

12241257
Controls whether failed workflow runs are grouped under a parent "[aw] Failed runs" issue. This is opt-in and defaults to `false`.

0 commit comments

Comments
 (0)