用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
直接命令不会经过审查 Prompt;运行前请先检查来源。
npx skills add https://github.com/jleechanorg/claude-commands --skill pr-preview-servers命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地?可下载 SkillsMP 当前能够提供的文件。
Token-efficient second opinion slash command /advice. Extracts decision point + artifact (≤150 lines), then fans out in parallel: (1) Opus subagent reviewer with fallback chain codex→agy→cursor, (2) /research on the decision topic, (3) /secondo multi-model opinion. Use instead of advisor() which ships the full conversation uncached.
Use this skill when working in repositories managed by Agent Orchestrator or when the user asks how to use `ao` properly. Covers the default AO workflow: bootstrap with `ao start`, dispatch work with `ao spawn`, inspect progress with `ao status` or `ao session ls`, steer sessions with `ao send`, and recover or clean up sessions safely. Includes strict parameter fidelity, pre-spawn cap cleanup, quota-wall fallback, and post-spawn verification.
Generate a full agento PR status report — draft readiness, canonical /green, zero-touch rate, inline display, and Slack summary.
正在显示 SKILL.md
基于 SOC 职业分类
| name | pr-preview-servers |
| description | Find and validate rotating GCP Cloud Run preview servers for PR branches. |
Reference for the rotating preview server pool used to deploy PR branches.
mvp-site-app-s1 through mvp-site-app-s10)mvp-site-appworldarchitecture-aius-central1$PROJECT_ROOT/**, Dockerfile, requirements.txt, etc..github/workflows/pr-preview.yml.github/scripts/pr-server-pool.shhttps://mvp-site-app-s{N}-{hash}-uc.a.run.app
Where {N} is 1–10 and {hash} is Cloud Run's stable unique URL segment.
From GitHub CLI (fastest):
gh pr view <PR_NUMBER> --json comments --jq '.comments[].body' | grep -o 'https://mvp-site-app-s[^ ]*'
From PR comments in the UI:
Look for the bot comment that says "Preview deployed to: https://..."
From Cloud Run directly:
gcloud run services describe mvp-site-app-s{N} \
--project=worldarchitecture-ai \
--region=us-central1 \
--format='value(status.url)'
The preview server is deployed with:
FLASK_ENV=productionENVIRONMENT=previewGEMINI_API_KEY, CEREBRAS_API_KEY, OPENROUTER_API_KEYTESTING_AUTH_BYPASS=true — real Firebase auth is required for protected endpointsGET /health
→ 200 {"status": "healthy", "service": "worldarchitect-ai", "timestamp": "..."}
Note: /health does not expose the ENVIRONMENT env var. Confirmed 2026-02-20.
Use the standalone smoke test (no local server startup needed):
# Test the current PR's preview server
python3 testing_mcp/test_openclaw_gateway_url_preview.py
# Test a specific preview URL
python3 testing_mcp/test_openclaw_gateway_url_preview.py \
--server-url https://mvp-site-app-s10-i6xf2p72ka-uc.a.run.app
| Test | Why |
|---|---|
/health → 200 | Public health endpoint |
/settings HTML → 200, field present | Public HTML page |
POST /api/settings → 401 (not 404) | Route wired, auth enforced |
| Static assets | Public static files |
POST /api/settings with actual settings values (save to Firestore)GET /api/settings (read user settings)When all 10 slots are taken, the oldest assignment is evicted to make room. The evicted PR loses its preview URL. Check PR #N's comments for current status.
| PR | Server | URL |
|---|---|---|
| #5662 | s10 | https://mvp-site-app-s10-i6xf2p72ka-uc.a.run.app |
| #5857 | s2 | https://mvp-site-app-s2-i6xf2p72ka-uc.a.run.app |
Update this table as PRs are deployed. URL stays stable as long as the pool slot isn't evicted.