docs(cookbooks): use OpenAI / non-Anthropic models via a gateway #5
Workflow file for this run
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| name: self-guard | |
| # agent-ops eating its own dog food: run THIS repo's own guard.yml hygiene | |
| # gate (the leakage linter) on agent-ops's own PRs + main pushes, via a local | |
| # reusable-workflow ref (`uses: ./...`). The leakage scan runs against this | |
| # PR's tracked files + this PR's scripts/lint/leakage-allowlist.txt; the linter | |
| # logic itself comes from guard.yml's pinned agent-ops checkout (default main), | |
| # and script breakage is separately caught by ci.yml's shellcheck. | |
| # | |
| # Zero infra: guard.yml runs on GitHub-hosted ubuntu and needs no secrets or | |
| # self-hosted runner. run_secret_checks stays false (agent-ops has no LITELLM / | |
| # AGENT_GH_TOKEN secrets, and check-secret-values is consumer-specific). | |
| # | |
| # The agentic surfaces (triage/review/implement/revise) are wired in | |
| # self-triage.yml etc. but stay dormant until secrets + a runner are | |
| # provisioned — see SELF-DOGFOOD.md. | |
| on: | |
| pull_request: | |
| push: | |
| branches: [main] | |
| permissions: | |
| contents: read | |
| jobs: | |
| guard: | |
| uses: ./.github/workflows/guard.yml | |
| with: | |
| run_secret_checks: false |