Help us improve
Share bugs, ideas, or general feedback.
From product-playbook-for-agentic-coding
Patterns for autonomous project execution with minimal human intervention. Use this skill when executing well-defined tasks autonomously, including validation strategies, stop conditions, and quality gates. Don't use when doing a single task interactively with the user, or when the project lacks a tasks document.
npx claudepluginhub daviswhitehead/product-playbook-for-agentic-coding-plugin --plugin product-playbook-for-agentic-codingHow this skill is triggered — by the user, by Claude, or both
Slash command
/product-playbook-for-agentic-coding:autonomous-executionThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
This skill provides patterns for executing projects autonomously while maintaining quality and knowing when to stop and ask for help.
Guides technical evaluation of code review feedback: read fully, restate for understanding, verify against codebase, respond with reasoning or pushback before implementing.
Share bugs, ideas, or general feedback.
This skill provides patterns for executing projects autonomously while maintaining quality and knowing when to stop and ask for help.
Before starting autonomous work, verify:
When to use: Standard execution for most projects
Process:
Validation Checkpoints:
When to use: Projects with clear phases (migrations, refactors)
Process:
When to use: Large projects or risky changes
Process:
Write session checkpoints — context compaction is inevitable in long runs, not an edge case:
docs/checkpoints/latest.mdSee the session-checkpoint skill for the full checkpoint format.
When running multiple independent tasks, consider whether to use git worktrees for isolation:
| Scenario | Recommendation | Reason |
|---|---|---|
| Sequential tasks in one subsystem | Stay on branch | Low isolation benefit, worktree overhead not worth it |
| Independent tasks across 2+ subsystems | Parallel worktrees | Tasks can't interfere with each other, enables parallel agents |
| Background maintenance tasks (lint fixes, doc updates) | Always use worktree | Keeps primary branch clean for feature work |
| Risky or experimental changes | Use worktree | Easy to discard without affecting main work |
When NOT to use worktrees:
After completing each task:
Run relevant tests:
Check acceptance criteria:
Run quality checks:
Update task status:
After completing each phase:
CRITICAL: Stop and ask for help when encountering:
| Condition | Action |
|---|---|
| Tests fail 3+ times | Stop, document issue |
| Architectural decision needed | Stop, ask for guidance |
| External dependency blocked | Stop, report blocker |
| Quality gate failure (can't fix) | Stop, seek help |
| Ambiguous requirements | Stop, ask for clarification |
| Same tool fails 2+ times | Switch strategy, don't brute-force |
| Unsure about platform capability | Say "I'm not sure, let me verify" — don't state as fact |
| Condition | Action |
|---|---|
| Unexpected complexity | Assess scope, consider asking |
| Breaking changes detected | Evaluate impact, may need input |
| Performance regression | Investigate, may need guidance |
| Security concerns | Always ask before proceeding |
When stopping:
Document current state:
Update task status:
Create summary:
Before each commit, discover and run project-specific validation (from CLAUDE.md / package.json):
ci:local > test:verify > test:pre-push > individual commandsnpm run ci:local): run it instead of individual commandsAt minimum verify:
Before proceeding to next phase:
Before marking project complete:
This skill supports:
/playbook:work - Primary execution command/playbook:tasks - Task document management/playbook:learnings - Post-execution captureAutonomous execution is successful when:
Autonomous execution requires discipline: validate often, stop when uncertain, document everything.