项目文件夹

文件
Asim Aslam 5f40cae7af ci: add DevRel + Architect overseer passes to the loop (#3085)
The hourly loop ships increments but nothing watches the whole. Add two
periodic high-altitude passes, same dispatch mechanism (fresh issue → @codex):

- devrel-review.yml (daily): audits README, website, docs, blog for coherence
  with the North Star, README crispness, and blog-worthy material. Safe
  alignment/crispness fixes auto-merge; brand/positioning copy and blog drafts
  are surfaced in a report for the human, never auto-merged.
- architecture-review.yml (every ~3 days): reviews the framework/harness against
  the thesis and files scoped follow-up issues that feed the increment loop. It
  does not make breaking/architectural changes itself.

Documented both in CONTINUOUS_IMPROVEMENT.md (Overseer passes).


Claude-Session: https://claude.ai/code/session_01CmdEY7pYmV5zzwCjNJ4ykL

Co-authored-by: Claude <noreply@anthropic.com>
2026-06-25 19:42:58 +01:00

49 行
3.3 KiB
YAML

name: Architecture Review
# Periodic high-altitude review of the whole framework and harness against the
# North Star (internal/docs/THESIS.md) — part of the autonomous loop
# (internal/docs/CONTINUOUS_IMPROVEMENT.md). Where DevRel watches the public
# story, the architect watches the system: API coherence, lifecycle gaps, and
# whether recent increments are converging on the thesis or sprawling.
#
# The architect's OUTPUT is an assessment plus scoped follow-up issues that feed
# the hourly increment loop — NOT large refactors. Breaking public-API and
# architectural changes stay with the human (see CONTINUOUS_IMPROVEMENT.md).
#
# Opens a fresh issue and dispatches Codex via CODEX_TRIGGER_TOKEN.
on:
workflow_dispatch: {}
schedule:
- cron: "0 8 */3 * *" # roughly every 3 days, 08:00 UTC (tunable)
permissions:
issues: write
concurrency:
group: architecture-review
cancel-in-progress: false
jobs:
dispatch:
runs-on: ubuntu-latest
steps:
- name: Open an architecture review issue and dispatch Codex
env:
GH_TOKEN: ${{ secrets.CODEX_TRIGGER_TOKEN || github.token }}
HAS_TRIGGER_TOKEN: ${{ secrets.CODEX_TRIGGER_TOKEN != '' }}
REPO: ${{ github.repository }}
RUN_NUMBER: ${{ github.run_number }}
run: |
if [ "$HAS_TRIGGER_TOKEN" != "true" ]; then
echo "CODEX_TRIGGER_TOKEN is not set — skipping (Codex ignores Actions-bot comments)."
exit 0
fi
ISSUE_URL=$(gh issue create --repo "$REPO" \
--title "Architecture review #$RUN_NUMBER" \
--body "Periodic architecture / harness review against the North Star in internal/docs/THESIS.md. Output: an assessment plus scoped follow-up issues for the increment loop.")
ISSUE_NUM="${ISSUE_URL##*/}"
echo "Opened issue #$ISSUE_NUM — dispatching Codex (Architect)."
gh issue comment "$ISSUE_NUM" --repo "$REPO" --body \
"@codex Act as the architect for go-micro. Review the overall framework and harness against the North Star in internal/docs/THESIS.md (services → agents → workflows as one runtime) and the roadmap in ROADMAP.md. Assess: API coherence and consistency across the core packages (agent, ai, flow, gateway/mcp, gateway/a2a, model, server, store, registry), gaps or missing pieces in the services → agents → workflows lifecycle, duplication or drift, and whether recent increments (scan recently merged PRs) are converging on the thesis or sprawling. Then: (A) post a concise architectural assessment as a comment on this issue (#$ISSUE_NUM) — strengths, the top risks/gaps, and a recommended direction for the next increments; (B) file concrete, scoped follow-up issues for the highest-value gaps so the hourly increment loop can pick them up — \`gh issue create --label codex --label enhancement --title \"<scoped task>\" --body \"<goal, scope, acceptance criteria>\"\` (each must be a single, self-contained, CI-verifiable chunk). Do NOT make breaking public-API or architectural changes yourself — your output is the assessment and the issues. A small, safe doc/comment correction may be a PR (\`git switch -c codex/architect-$ISSUE_NUM\` … \`gh pr create --base master --label codex …\` … \`gh pr merge --squash --auto --delete-branch\`). Do not use the make_pr tool (it is a no-op stub)."