claude-harnesses

Everything at once: every skill, command, agent, and hook from all packs. Install this or individual packs, not both.

16 commands · 6 agents

"Run a continuous product improvement loop: evaluate a product or codebase, present prioritized recommendations for approval, then implement only approved improvements through verification and PR fol…

7 commands · 2 agents

Run an end-to-end Claude Code delivery workflow from user request to merge-ready pull request, including planning, implementation, tests, self-review, ready-for-review PR creation, CI/review follow-u…

6 commands · 2 agents

Watch PRs after creation: monitor CI, dispatch fixes via subagents, address review feedback.

3 commands · 1 agent

Product-quality workflows for frontend design, image-guided UI direction, kaizen evaluation loops, and merge-ready implementation delivery.

4 commands

Severity-ranked code review and multi-pass security review with parallel specialist subagents.

2 commands · 2 agents

Drive development with red-green-refactor using isolated test-writer, implementer, and refactorer subagents so tests reflect intent rather than anticipated implementation.

1 command · 3 agents

Red-Green-Refactor TDD with subagent context isolation between phases.

1 command · 3 agents

Compatibility alias for pr-guardian. Use pr-guardian for stalled pull requests, CodeRabbit/Codex review feedback, CI failures, unresolved review conversations, and merge-ready PR follow-up.

1 command · 1 agent

Diagnose and repair failing CI checks on the current PR by parsing logs, reproducing locally when possible, and pushing focused fixes in batches.

1 command · 1 agent

Monitor a pull request after opening it or resume a stalled PR, fix CI failures and reviewer feedback, push updates, reply to and resolve review conversations, and comment with the outcome.

1 command · 1 agent

Review code changes with a bug-first, risk-focused workflow before merge or release.

1 command · 1 agent

Prepare a verdict-first, read-only HTML briefing that helps a human review a pull request quickly and well. Use with a PR number or URL; opens in the browser with a terminal fallback.

1 command · 1 agent

Run a multi-pass security review over the current diff using parallel specialist subagents focused on injection, authn/authz, secrets, supply-chain, and infrastructure-as-code risks.

1 command · 1 agent

Fallback Adviser consultations using a review-only Opus subagent when the native Advisor tool is unavailable.

1 command

Survive context loss across sessions: SessionStart context injection, cost-ceiling guard, large-edit gate, ledger templates, goal-manager workflow.

1 command

Inspect recent Claude Code activity and promote repeated patterns into the smallest useful Skill, subagent, or hook. Read-only investigation; creation gated by user approval.

1 command

Research helpers for reading public web pages that normal browsing, scraping, snippets, or direct fetch tools cannot read cleanly.

1 command

Route an independent, review-only Claude process to a deterministically stronger model or effort when native Advisor is unavailable.

1 command

Explicitly keep a GitHub pull request moving until it is merge-ready by delegating to PR Guardian. Invoke manually with /autopilot; Claude must not start it automatically.

1 command

Guidance for distinctive, intentional visual design when building new UI or reshaping an existing one.

1 command

Fetch public web pages with Jina AI Reader API and convert them into LLM-friendly Markdown.

1 command

Analyze recent Claude Code work and package repeated patterns as the smallest useful skill, custom subagent, slash command, hook, or automation.

1 command

Teach the human everything that happened in this session until they can demonstrably explain it, using an incremental checklist, restate-first probing, and AskUserQuestion quizzes.

1 command

Direct high-quality frontend UI work by using image-generated mockups when useful, translating the chosen direction into code, and verifying the rendered UI in a browser.

1 command

Teach the human everything that happened in a session until they can demonstrably explain it, via an incremental checklist and AskUserQuestion quizzes.

1 command

Verification feedback loop: format/typecheck/test on every edit, and block premature Stop until verify.sh passes.

1 command

Curated .mcp.json recipe for GitHub, Playwright, Context7, Serena, Sequential-Thinking, Sentry, and Excalidraw. All servers ship disabled.

Deterministic PreToolUse guards: secret detection, dangerous commands, branch protection, prompt injection, MCP allowlisting.

Fix bugs with reproduction, minimal changes, regression coverage, and targeted verification.

Remove AI-generated boilerplate (vague comments, unused try/except, defensive validation, restated docstrings) from changed code without changing behavior.

Update documentation after code or behavior changes while keeping docs concise and relevant.

Iteratively improve a skill or prompt by having a fresh subagent execute it against fixed scenarios, scoring two-sided (executor self-report + a pre-frozen requirements checklist), and looping until…

Implement a feature safely with planning, focused edits, tests, docs, and verification.

Manage durable task goals for long-running Claude Code work using the project ledger, including objective writing, success criteria, resumption, and completion decisions.

Run multi-hour or multi-session work safely by combining the ledger, checkpoints, cost-ceiling guard, and resumable progress notes so context loss does not erase progress.

MODIFIER skill that removes AI slop from prose without losing source facts or uncertainty. Use with any writing task or to rewrite text; for code slop, use deslop instead.

Refactor code without behavior changes using small mechanical steps and verification.

Check whether a change is ready to release by reviewing diff, tests, docs, risks, and migration notes.

Review changed code on the current branch for reuse, clarity, and dead complexity, then propose or apply minimal cleanup without changing behavior.