skills/ancplua/workflow-orchestration/SKILL.md
Coordinates multi-step CI/CD pipelines by chaining autonomous-ci, code-review, smart-commit, and jules-integration plugins. Use when executing validation-to-PR workflows or recovering from CI failures.
npx skillsauth add aiskillstore/marketplace workflow-orchestrationInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
This skill orchestrates complex workflows by chaining existing plugins into pipelines. It maintains state awareness through CHANGELOG.md and coordinates cross-plugin operations.
Required Plugins:
autonomous-ci - Validation and CI monitoringcode-review - Quality analysissmart-commit - Commit message generationOptional Plugins:
jules-integration - Async PR delegationRequired Skills (from Superpowers):
systematic-debugging - For failure analysisverification-before-completion - For evidence-based completionBefore executing any pipeline:
# 1. Read CHANGELOG for recent context
Read the file: CHANGELOG.md
# Focus on [Unreleased] section to understand pending work
# 2. Check git state
git status --porcelain
git log --oneline -5
# 3. Verify required plugins are available
claude plugin list
Choose the appropriate pipeline based on the task:
| Task | Pipeline | Plugins Used |
|------|----------|--------------|
| Validate before commit | pre-commit | autonomous-ci, code-review, smart-commit |
| Create PR | pr-create | All + jules-integration |
| Fix CI failure | ci-recover | autonomous-ci + systematic-debugging |
Step 1: Validate
├── Run: ./tooling/scripts/local-validate.sh
├── On success: Continue
└── On failure: STOP, report issues
Step 2: Review
├── Invoke: code-review skill
├── Analyze: Security, style, performance
├── On success: Continue
└── On failure: Report issues, suggest fixes
Step 3: Commit Message
├── Invoke: smart-commit skill
├── Generate: Conventional commit message
├── On success: Present message for human approval
└── On failure: Provide manual template
Step 1: Pre-Commit Pipeline
├── Execute full pre-commit pipeline
└── On failure: STOP
Step 2: Stage Changes
├── Present diff for human review
├── REQUIRE: Human approval for git commit
└── On approval: Continue
Step 3: Delegate to Jules
├── Invoke: jules-integration skill
├── Create: Jules session with PR task
├── Set: requirePlanApproval = true
└── Monitor: Session status
Step 1: Diagnose
├── Invoke: systematic-debugging skill
├── Analyze: CI failure logs
├── Identify: Root cause
└── On failure: Escalate to human
Step 2: Fix
├── Implement fix based on diagnosis
├── Follow TDD if adding code
└── On failure: Escalate with context
Step 3: Re-validate
├── Run: ./tooling/scripts/local-validate.sh
├── On success: Report recovery
├── On failure: Retry (max 3 times)
└── On max retries: Escalate
ALWAYS verify before claiming complete:
# Run validation
./tooling/scripts/local-validate.sh
# Check all tests pass
# (if applicable)
# Update CHANGELOG.md under [Unreleased]
Report with evidence:
## Pipeline Execution Summary
| Step | Status | Evidence |
|------|--------|----------|
| Validation | ✅ | local-validate.sh passed |
| Review | ✅ | No critical issues |
| Commit | ⏳ | Awaiting human approval |
### Artifacts
- Commit message: `feat: add workflow orchestration`
- Files changed: 5
- Lines: +230 / -12
The orchestrator reads state from multiple sources:
# Extract pending work
grep -A 50 "\[Unreleased\]" CHANGELOG.md
Provides:
git status --porcelain # Working directory state
git log --oneline -5 # Recent commits
git diff --stat # Changes summary
# Via autonomous-ci skill
./plugins/autonomous-ci/scripts/wait-for-ci.sh
| Skill | When Used |
|-------|-----------|
| systematic-debugging | CI failure diagnosis |
| test-driven-development | Writing fixes |
| verification-before-completion | Before claiming done |
| brainstorming | Complex architectural decisions |
| Skill | When Used |
|-------|-----------|
| autonomous-ci | Validation and monitoring |
| code-review | Quality analysis |
| smart-commit | Commit generation |
| jules-integration | PR delegation |
| working-on-ancplua-plugins | Repo conventions |
Trigger: Developer requests validation before commit
1. Read CHANGELOG.md → Understand context
2. Run local-validate.sh → All checks pass
3. Invoke code-review → No critical issues
4. Invoke smart-commit → Generate: "feat(agent): add workflow orchestrator"
5. Present summary → Human approves
6. Report complete with evidence
Trigger: CI fails on shellcheck
1. Detect failure via autonomous-ci monitoring
2. Invoke systematic-debugging:
- Phase 1: Gather evidence (CI logs)
- Phase 2: Identify cause (SC2086 unquoted variable)
- Phase 3: Hypothesize fix (add quotes)
- Phase 4: Verify fix locally
3. Apply fix to script
4. Run local-validate.sh → Passes
5. Report recovery with evidence
Trigger: Feature complete, ready for review
1. Execute pre-commit pipeline → All green
2. Present diff for human review
3. Human approves commit
4. Invoke jules-integration:
- Create session: "Create PR for workflow-orchestrator agent"
- Set requirePlanApproval: true
- Monitor session
5. Report PR URL when created
development
Apple Human Interface Guidelines for content display components. Use this skill when the user asks about charts component, collection view, image view, web view, color well, image well, activity view, lockup, data visualization, content display, displaying images, rendering web content, color pickers, or presenting collections of items in Apple apps. Also use when the user says how should I display charts, what's the best way to show images, should I use a web view, how do I build a grid of items, what component shows media, or how do I present a share sheet. Cross-references: hig-foundations for color/typography/accessibility, hig-patterns for data visualization patterns, hig-components-layout for structural containers, hig-platforms for platform-specific component behavior.
tools
Automate HelpDesk tasks via Rube MCP (Composio): list tickets, manage views, use canned responses, and configure custom fields. Always search tools first for current schemas.
testing
Expert Haskell engineer specializing in advanced type systems, pure functional design, and high-reliability software. Use PROACTIVELY for type-level programming, concurrency, and architecture guidance.
tools
GraphQL gives clients exactly the data they need - no more, no less. One endpoint, typed schema, introspection. But the flexibility that makes it powerful also makes it dangerous. Without proper controls, clients can craft queries that bring down your server. This skill covers schema design, resolvers, DataLoader for N+1 prevention, federation for microservices, and client integration with Apollo/urql. Key insight: GraphQL is a contract. The schema is the API documentation. Design it carefully.