Repository navigation
[docs] docs: unbloat FAQ answers - #63061
Conversation
|
✅ Test Quality Sentinel completed test quality analysis. No test files were added or modified in this PR. Test Quality Sentinel skipped. Warning Firewall blocked 1 domainThe following domain was blocked by the firewall during workflow execution:
To allow these domains, add them to the network:
allowed:
- defaults
- "o205451.ingest.us.sentry.io"See Network Configuration for more information.
|
There was a problem hiding this comment.
Note
Copilot could not run the full agentic suite for this review because it was automatically requested on a bot-authored pull request. Request a review from Copilot under Reviewers to retry with the full agentic suite. Improved support for bot-authored pull requests is coming soon.
Copilot review overview
Review effort: Lite
Findings: 6
Open (6)
This sentence has a comma splice (“..., do not ...”) and is harder to read/translate. Split into… · New Several answers are now single very-long lines. This makes future edits produce noisy diffs and is… · New This rewrite drops the previously documented permission requirements per option (e.g., `issues:… · New Several answers are now single very-long lines. This makes future edits produce noisy diffs and is… · New The text references bothmax-turnsandmax-continuations, but the example only shows… · New Several answers are now single very-long lines. This makes future edits produce noisy diffs and is… · New
What changed in this PR
This PR streamlines the FAQ reference documentation by converting several multi-line/bulleted explanations into single-paragraph prose.
Changes:
- Condensed multiple bulleted guidance sections (security, lock files, PR fallback behavior, billing/cost controls) into one-paragraph summaries.
- Adjusted the “create workflow” prompt example formatting to be presented outside of a bulleted list.
| File | Description |
|---|---|
| docs/src/content/docs/reference/faq.md | Condenses several FAQ answers and guidance blocks into shorter, single-paragraph text. |
💡 Add a code-review agent skill for context-aware, tailored reviews. Learn more in the docs.
| - Keep keys specific to OS and dependency lock state (`go.sum`) to reduce accidental cross-context restores. | ||
| - Do not share writeable cache keys across trust boundaries (for example, untrusted fork PR runs and protected branch runs). | ||
| - Never place secrets in `GOMODCACHE`/`GOCACHE`; these directories should contain only modules and build outputs. | ||
| Keep cache keys specific to the OS and `go.sum`, do not share writable keys across trust boundaries such as untrusted forks and protected branches, and never place secrets in `GOMODCACHE` or `GOCACHE`. |
|
|
||
| - **`.md`**: source; edit the prompt body freely without recompiling | ||
| - **`.lock.yml`**: what GitHub Actions runs; regenerate after any frontmatter change | ||
| The `.lock.yml` file is the compiled GitHub Actions workflow generated from your `.md` by `gh aw compile`. It contains SHA-pinned actions, resolved imports, permissions, and all guardrail hardening, so inspect it to see exactly what will run. Commit both files: the `.md` is the editable source, while the `.lock.yml` is what GitHub Actions runs and must be regenerated after any frontmatter change. |
| - **Disable fallback**: set `fallback-as-issue: false` to fail when PR creation is blocked (requires only `contents: write` and `pull-requests: write`). | ||
|
|
||
| See [Pull Request Creation](/gh-aw/reference/safe-outputs/#pull-request-creation-create-pull-request). | ||
| Some organizations block PR creation by GitHub Actions (**Settings → Actions → General → Workflow permissions**). If you can't enable it, keep the default `fallback-as-issue: true` to open an issue with the branch link instead, assign that issue to `copilot` if you want automated follow-up (`assignees: [copilot]` under `create-issue`), or set `fallback-as-issue: false` to fail outright. See [Pull Request Creation](/gh-aw/reference/safe-outputs/#pull-request-creation-create-pull-request). |
| It depends on the engine: GitHub Copilot CLI bills the organization through `copilot-requests: write` or the account supplying [`COPILOT_GITHUB_TOKEN`](/gh-aw/reference/auth/#copilot_github_token); Claude Code bills the Anthropic account tied to [`ANTHROPIC_API_KEY`](/gh-aw/reference/auth/#anthropic_api_key) or Anthropic WIF; OpenAI Codex bills the OpenAI account tied to `CODEX_API_KEY` or [`OPENAI_API_KEY`](/gh-aw/reference/auth/#openai_api_key); Google Gemini bills the Google account tied to [`GEMINI_API_KEY`](/gh-aw/reference/auth/#gemini_api_key) or Google WIF; and Pi bills the provider selected in the `provider/model` value. | ||
|
|
|
|
||
| - `max-turns` (Claude) — limits AI chat iterations per run | ||
| - `max-continuations` (Copilot) — autopilot mode with consecutive triggered runs | ||
| gh-aw has no automatic retries, so each trigger produces one run. Bound tokens and wall-clock time with `max-turns` for Claude or `max-continuations` for Copilot: |
| Spend controls live at the provider level: use GitHub Billing for Actions minutes, provider-side API key or project limits for Claude, Codex, and Gemini, and Copilot plan quota limits for Copilot. For per-repository tracking, use a dedicated API key per repository, `gh aw audit <run-id>` for per-run detail, and `gh aw logs` for aggregate metrics. | ||
|
|
|
🎉 This pull request is included in a new release. Release: |

Summary
docs/src/content/docs/reference/faq.mdBloat removed
Impact
docs/src/content/docs/reference/faq.md