القائمة
Fetch a GitHub issue and run the full grooming → implementation → review → QA pipeline.
التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.
استنادا إلى تصنيف SOC المهني
Generate a bespoke issue-workflow for any target project. Phase 1 — a Claude Opus analyst reads the project deeply and produces a disposition report. Phase 2 — on approval, parallel writer agents transplant every Maestro workflow component, adapted to the target project's actual stack, test runner, dev environment, and conventions. Use when a project needs its own self-contained issue-workflow without taking a direct dependency on Maestro.
Show the Maestro command map — a quick reference to all available skills.
Adversarially review a grooming spec before implementation starts. Finds hidden risks, unvalidated assumptions, and missing dependencies. Standalone entry point for the challenger agent.
Run QA validation on a pull request — boots the local environment, tests acceptance criteria, and optionally posts the report as a PR comment. Standalone entry point for the qa-engineer agent.
Run a lead code review on the current branch or a given PR. Standalone entry point for the lead-reviewer agent.
Groom a single GitHub issue — produce an implementation spec and optionally post the grooming summary as a GitHub comment. Standalone entry point for the grooming-agent.
| name | issue-workflow |
| description | Fetch a GitHub issue and run the full grooming → implementation → review → QA pipeline. |
Read .claude/maestro.json to resolve project-specific values:
{TEMP_ROOT} = .ai.temp_root{REPO} = .ai.repo{SLUG} = .ai.slug{E2E_URL} = .ai.e2e.local_url{E2E_BOOT} = .ai.e2e.boot_cmd{E2E_SETTINGS} = .ai.e2e.settings_path{E2E_CI} = .ai.e2e.ci_integrationRepository: {REPO} (from .claude/maestro.json)
When the user asks to work on an issue by number, such as:
/task 123issue 123#123follow this workflow. The orchestrator runs inline in this conversation — read the user's opening message before kicking it off, since it uses that for escalation calibration (high autonomy / standard / high oversight).
Use shell commands as the primary approach. The GitHub MCP (mcp_github_*) may be used if it is connected — but shell is always the safe fallback and is preferred for reliability.
| Operation | Shell (primary) | GitHub MCP (if connected) |
|---|---|---|
| Issue fetch | bash .claude/skills/issue-workflow/scripts/issue-sync.sh <N> | mcp_github_github_issue_read |
| Branch creation | bash .claude/skills/issue-workflow/scripts/make-issue-branch.sh | — |
| Staging & committing | git add / git commit | — |
| Pushing | git push -u origin <branch> | — |
| PR creation | gh pr create | mcp_github_github_create_pull_request |
| CI monitoring | gh pr checks <PR#> | mcp_github_github_pull_request_read |
Extract the issue number from the user's message.
Fetch the issue — run bash .claude/skills/issue-workflow/scripts/issue-sync.sh <N> (or use the MCP equivalent). Read the resulting file at {TEMP_ROOT}/issues/<N>/issue.md.
Check for parent epics — if Parent Epic (GitHub) or Parent Epics (Task List) has entries, sync each parent with issue-sync.sh <epic-N> and read those files at {TEMP_ROOT}/issues/<epic-N>/issue.md for context.
Check if this is an Epic — if the issue has label epics, Issue Type EPIC, or has sub-issues listed, ask the user: "Work the epic as a whole, or a specific sub-issue?" If a sub-issue is chosen, sync it and proceed with the epic context in mind.
Determine base branch — default is origin/develop unless the user specified otherwise.
Invoke the orchestrator skill inline (do not spawn it as a sub-agent — it runs in this conversation context so it can read the user's intent for escalation calibration):
Inputs: issue number
N, issue file{TEMP_ROOT}/issues/<N>/issue.md, base branch
The orchestrator skill manages everything from here: calibration → grooming → spec review → dispatch → implementation → lead review → push & PR → CI → QA → finalize. It spawns the specialist agents (grooming-agent, challenger, backend-agent, frontend-agent, release-agent, lead-reviewer, qa-engineer, ticket-writer) as isolated sub-agents, but the orchestrator itself stays inline so it can surface decisions back to the user naturally.
Monitor progress at {TEMP_ROOT}/issues/<N>/workflow-log.html.