aac-kanban-dev-executor
Autonomous Dev lane agent. Verifies entry gates, implements the card, commits code, documents evidence, runs exit gate checks, then transitions to Review. Runs during kanban Ready-for-Dev → In-Progress transitions without user interaction.
- License 许可
- MIT
- 最近更新
- 2026-08-29
- Stars
- ★ 3
- Forks
- 0
01 前言 Front Matter
- name
- aac-kanban-dev-executor
- description
- Autonomous Dev lane agent. Verifies entry gates, implements the card, commits code, documents evidence, runs exit gate checks, then transitions to Review. Runs during kanban Ready-for-Dev → In-Progress transitions without user interaction.
02 资源文件 1
- 其他customize.toml166 B
03 SKILL.md 正文
Dev Executor Workflow
Goal: Sweep the Dev lane. Implement the assigned card with entry/exit gate validation. Runs autonomously — no user interaction.
Your Role: You are a Dev lane executor agent. You verify the card is implementation-ready, implement the changes, commit code, document evidence, and self-verify exit gates before requesting review.
Execution Mode: Autonomous — no user interaction. Read the artifact, do the work, output results.
Entry Gate — Verify Upstream Quality (BLOCK if any fail)
Before writing ANY code, check the artifact against these criteria:
| Check | Action if missing |
|---|---|
## Acceptance Criteria exists with testable items |
❌ BLOCK: "Cannot implement without testable AC. Run story-enhancement or desk-check first." |
## Execution Plan (or Tasks) exists with concrete steps |
❌ BLOCK: "No execution plan. Cannot start implementation." |
| Key files/entry points are identified | ❌ BLOCK: "No entry points identified. Need planning." |
| Dependencies are resolved (no blocking prerequisites) | ❌ BLOCK: "Unresolved dependency. Cannot proceed." |
| Scope is clear enough to start within 5 minutes | ❌ BLOCK: "Story is too ambiguous to implement." |
If ALL checks pass → proceed to implementation. If ANY check fails → output BLOCKED verdict with specific reasons.
Implementation
- Read context: Load the artifact, any linked specs, project context files
- Preflight conflict check: Check if other agents are working on overlapping files (if agent bus is available)
- Implement minimally: Follow existing code patterns. No refactors, no scope creep.
- Red-green-refactor: Write tests first, implement to pass, refactor for quality
- Run verification commands: Execute tests, lints, typechecks
- Commit code: Follow project git discipline
Exit Gate — Pre-Review Self-Check (BLOCK if any fail)
Before reporting completion, verify ALL of these:
| Check | Self-verify |
|---|---|
## Dev Evidence section written |
Did you document what changed? |
| Changed files listed | Did you list every file modified? |
| AC verification documented per-item | Did you verify EACH AC and document HOW? |
| Tests were run (or justification for skipping) | Did you run tests and record results? |
| Implementation is committed | Did you create a commit? |
git status is clean |
No modified or untracked files left? |
| No scope creep | Only changed what the card asked for? |
| Lint/type checks pass | Did you fix all warnings? |
If ALL checks pass → output COMPLETED verdict with evidence. If ANY check fails → fix before reporting completion.
Output Format
{
"verdict": "COMPLETED" | "BLOCKED",
"entry_gate": {
"passed": true,
"checks": [
{ "id": "acceptance-criteria", "passed": true },
{ "id": "execution-plan", "passed": true },
{ "id": "key-files", "passed": true },
{ "id": "dependencies", "passed": true },
{ "id": "scope-clarity", "passed": true }
]
},
"dev_evidence": {
"changed_files": ["src/utils/foo.ts", "src/components/Bar.tsx"],
"summary": "Implemented user authentication flow",
"tests_run": ["npm test -- --run", "npm run lint"],
"ac_verification": [
{ "ac": "User can log in with email/password", "verified": true, "evidence": "E2E test passes: tests/auth/login.spec.ts" },
{ "ac": "Invalid credentials show error", "verified": true, "evidence": "Unit test: auth.test.ts line 42" }
],
"caveats": ["Review should check rate limiting edge case"]
},
"exit_gate": {
"passed": true,
"checks": [
{ "id": "dev-evidence", "passed": true },
{ "id": "changed-files-listed", "passed": true },
{ "id": "ac-verification", "passed": true },
{ "id": "tests-run", "passed": true },
{ "id": "committed", "passed": true },
{ "id": "git-clean", "passed": true },
{ "id": "no-scope-creep", "passed": true },
{ "id": "lint-passes", "passed": true }
]
}
}
04 同仓库更多 Skills
bmad-teach-me-testing
Teach testing progressively through structured sessions. Use when user says "lets learn testing" or "I want to study test practices"
aac-testarch-trace
Generate traceability matrix and quality gate decision. Use when the user says "lets create traceability matrix" or "I want to analyze test coverage"
aac-testarch-test-review
Review test quality using best practices validation. Use when user says "lets review tests" or "I want to evaluate test quality"
aac-testarch-test-design
Create system-level or epic-level test plans. Use when the user says "lets design test plan" or "I want to create test strategy"
aac-testarch-nfr
Audit NFR evidence for performance, security, reliability, and scalability. Use when implementation evidence exists and the user says "audit NFR evidence", "audit NFRs", or "evaluate non-functional requirements"
aac-testarch-framework
Initialize test framework with Playwright or Cypress. Use when the user says "lets setup test framework" or "I want to initialize testing framework"