CLAUDE CODE PLAYBOOKS

Reshape PR Commits

Turn a branch's messy WIP/fixup commit history into a clean, logically-grouped set of commits before review — proposed regrouping with explicit approval, git reset --soft plus staged re-commits instead of interactive rebase, and hard guardrails (feature branches only, force-with-lease only, never touch another author's commits).

Academic Research

Scholarly Writing Refiner

Polish academic English at the sentence level — a five-dimension review (grammar, word choice, voice, coherence, sentence structure) with concrete before/after tables for colloquial-to-academic substitutions, active/passive voice selection by scenario, transition-signal words by logical relationship, and section-specific tense and structure conventions for the Abstract and Introduction.

Planning with Files

Persistent, file-based planning for multi-step AI-agent work — task_plan.md, findings.md, and progress.md on disk, a 2-action rule for capturing multimodal findings before they're lost, a 3-strike error protocol, and a 5-question reboot test to verify state survives a compaction.

PR Queue Triage

Clear a backlog of open pull requests before a release by classifying every PR into an evidence-based disposition — never by title — with a real git merge-tree test against the actual release branch, not the platform's often-wrong mergeable flag.

Project Management

Project Status Artifact

Generate and publish an opinionated, tabbed project status page for work too big for one update — success criteria, a workstream sequence, next steps, risks — then refresh it in place on the same URL, reporting only what changed instead of re-narrating the whole project.

Reproducible Database Lookup

A methodology for querying public database APIs — scientific, regulatory, financial, or otherwise — so another agent or human can repeat exactly what you did: bounded calls, count reconciliation, identifier-conversion tracking, and untrusted-data handling for every response.

Project Graveyard: Autopsy Your Abandoned Side Projects

Scan local repos for dead side projects, autopsy each one from its git history, surface your personal death patterns, and pick the one corpse most worth resurrecting — then help ship it.

Media Production

Podcast Production Team

A coordinated five-role agent team — researcher, scriptwriter, show-note editor, distribution manager, and reviewer — that produces a full podcast episode's content in one pass, with the team size scaling to match the actual request.

PR Review Toolkit

Six specialist reviewers — comments, tests, error handling, type design, general quality, and simplification — each triggered by name or automatically based on what changed in the diff.

Ralph Wiggum Autonomous Loop

Self-referential development loop that keeps Claude iterating on the same task until it hits a completion promise or an iteration cap — a Stop hook that blocks exit and re-feeds the same prompt.

Redesign Existing Projects: UI Audit and Upgrade

A design audit checklist that finds generic AI-look patterns in an existing codebase and fixes them without breaking functionality or migrating frameworks

Creative & Design

PPT Creator

Create professional slide decks from topics or documents — structured content with data-driven charts, speaker notes, and complete PPTX files, applying the Pyramid Principle and assertion-evidence storytelling.

Product Management

Product Analysis

Multi-path parallel product analysis with cross-model test-time compute — spawn parallel agents to explore your product from multiple angles, then synthesize an actionable optimization plan.

20 minutes

Prompt Optimizer (EARS)

Transform vague prompts into precise, well-structured specifications using EARS (Easy Approach to Requirements Syntax) — ideal for AI-generated code, products, and docs.

Promptfoo Evaluation

Configure and run LLM evaluations with Promptfoo — build promptfooconfig.yaml, write Python custom assertions, implement llm-rubric judges, and compare models systematically.

QA Expert

Establish a comprehensive QA process — test strategies, Google-standard test cases, P0–P4 bug tracking, quality metrics, OWASP security testing, and 90% coverage targets.

20 minutes

Repomix Safe Mixer

Package codebases with repomix while automatically detecting and removing hardcoded credentials before packing — so you never leak secrets when sharing code.

Repomix Unmixer

Extract files from repomix-packed repositories — restore the original directory structure from XML, Markdown, or JSON output, reversing the packing process.

Research & Writing

Personal Research Analyst

Deliver structured, sourced, six-section research briefings every time — executive summary, key findings with citations, context, conflicting views, implications, and a ranked source list.

Productivity

AI Prompt Builder

Turn rough ideas into optimized, high-performance prompts for any AI model — eliminates ambiguity, adds the right context, and structures output expectations so you get better results on the first try.

Popup CRO

Create popups, modals, slide-ins, and banners that convert without annoying users or damaging brand.

Product Marketing Context

Create and maintain a foundational product marketing context document that other marketing skills reference automatically.

Referral & Affiliate Programs

Design and optimize referral, affiliate, and ambassador programs that turn customers into growth engines.

RevOps Playbook: Lead Scoring, Routing & MQL-to-SQL Handoff

Use Claude Code to design lifecycle stages, a fit-plus-engagement lead scoring model, routing rules with fallbacks, MQL/SQL handoff SLAs and CRM hygiene rules, built from your own closed-won data.

20 minutes