Guides an agent to find and fix wasteful work in REA binary-analysis tools, verified by before/after checks.
morluto/rea
Prepares REA release changelogs and GitHub release notes with verified contributor credits.
morluto/rea
Audits REA's CI workflows and test suites for false passes, fixture drift, and safe speedups
morluto/rea
Guides an agent in designing or changing REA investigation tools and their CLI/MCP contracts
morluto/rea
Lets an agent analyze shipped binaries, packages, and apps with REA tooling and cite evidence
morluto/rea
Makes the agent ask clarifying questions and agree on a design before building anything new.
obra/superpowers
Investigates a failed or slow agent session from its transcripts and writes a cited report
obra/superpowers
Split several independent problems across isolated sub-agents that work at the same time
obra/superpowers
Lets an agent carry out a written implementation plan itself, task by task, with a progress ledger and tests
obra/superpowers
Guides an agent through merging, pushing as a PR, or keeping finished git work, with tests checked first.
obra/superpowers
Makes the agent verify code review feedback technically before acting on it, instead of agreeing blindly
obra/superpowers
Gets an agent to hand finished work to a separate reviewer subagent before merging
obra/superpowers
Runs a multi-task plan by delegating each task to fresh AI subagents with automatic review
obra/superpowers
Makes the agent find the root cause of a bug before proposing any fix.
obra/superpowers
Makes the agent write a failing test before any feature, bugfix, or behavior change code.
obra/superpowers
Sets up an isolated git workspace for feature work so the agent doesn't disturb your current branch.
obra/superpowers
Makes the agent check for and invoke relevant skills before it responds or acts
obra/superpowers
Makes the agent run real checks and show evidence before saying work is done or passing
obra/superpowers
Turns a spec or requirements into a step-by-step, bite-sized implementation plan for an engineer.
obra/superpowers
Guides an agent to create, edit, and test skills using a test-driven documentation process
obra/superpowers
Routes you to the right Matt Pocock skill or workflow for your situation
mattpocock/skills
Reviews your branch's changes against repo coding standards and the original spec, side by side.
mattpocock/skills
Guides the agent to design deep modules with small interfaces, clean seams, and testable boundaries.
mattpocock/skills
Guides the agent through a disciplined loop to reproduce, narrow down and fix hard bugs
mattpocock/skills
Helps an agent build and sharpen a project's shared glossary and architecture decision records
mattpocock/skills
Interviews you about a plan or design, then writes ADRs and a glossary as you go
mattpocock/skills
Builds a spec or ticket end to end, using TDD, typechecks, tests, a review pass, and a commit.
mattpocock/skills
Builds a full spec in code by running ticket-based implementation with parallel subagents and reviews.
mattpocock/skills
Finds shallow modules and refactor opportunities in a codebase, shows them in an HTML report, then helps pick one.
mattpocock/skills
Writes pull request descriptions in a fixed Summary, Evidence and Merge Danger format.
mattpocock/skills
Builds quick disposable prototypes to test logic/state models or compare UI variations
mattpocock/skills
Delegates a research question to a background agent that saves cited findings as a Markdown file
mattpocock/skills
Reviews a past coding session and suggests fixes to the agent's setup (docs, checks, rules).
mattpocock/skills
Configures a repo's issue tracker, triage labels and domain docs for the engineering skills
mattpocock/skills
Guides an agent through red-green-refactor, writing tests test-first at agreed seams.
mattpocock/skills
Turns the current conversation into a structured spec and publishes it to your project issue tracker.
mattpocock/skills
Splits a plan or conversation into dependency-linked tickets and publishes them to a tracker or local files.
mattpocock/skills
Lets an agent triage GitHub-style issues and external PRs: label, verify, ask follow-ups, and write agent briefs.
mattpocock/skills
Plans large work as a shared map of decision tickets, resolved one at a time until the route is clear.
mattpocock/skills
Builds a guided bash script that walks you through manual setup steps like API keys and CI secrets.
mattpocock/skills
Runs a long-running goal across subagents, keeping a dated wiki of decisions and lessons.
mattpocock/skills
Hand the current chat to a fresh background Claude agent that continues the work.
mattpocock/skills
Interviews you to turn recurring tasks into delegable workflow specs saved as markdown files.
mattpocock/skills
Installs dependency-cruiser rules so TypeScript packages only expose their root entry-point files.
mattpocock/skills
Gives the agent a shared vocabulary and checklist for designing AI agent pipelines that turn tickets into merged code
mattpocock/skills
Turns a pile of raw notes into an article written one grounded beat at a time, with you choosing the path
mattpocock/skills
Mines raw writing fragments into a markdown file during an interview, before any structure is imposed
mattpocock/skills
Turns a raw pile of notes into a structured article, one agreed paragraph at a time.
mattpocock/skills
Blocks risky git commands like push, reset --hard and branch -D before Claude runs them.
mattpocock/skills
Swap risky `as` type casts in TypeScript tests for type-safe shoehorn helpers.
mattpocock/skills
Creates course exercise folder structures with stub readmes that pass the course linter
mattpocock/skills
Adds Husky and lint-staged pre-commit hooks that format, typecheck and test code before each commit.
mattpocock/skills
Interviews you relentlessly to sharpen a plan or design before you commit to it.
mattpocock/skills
Interviews you in rounds about a plan or idea until every branch of the decision tree is settled
mattpocock/skills
Summarises the current chat into a file a fresh AI agent can pick up and continue.
mattpocock/skills
Turns the agent into a stateful tutor that builds mission-driven HTML lessons and tracks your learning over sessions.
mattpocock/skills
Turns an unanswerable decision into a Markdown questionnaire for someone who has the answers.
mattpocock/skills
Re-explains the last message in plain, simplified English with context and project terminology
mattpocock/skills
Guides writing skills, AGENTS.md and CLAUDE.md files so agents follow them reliably.
mattpocock/skills
Guides an AI agent to diagnose its own repeated failures before retrying, and to write a structured debug report.
affaan-m/ECC
Trims an ECC install to the skills, rules and hooks a specific repo actually needs, using repo evidence.
affaan-m/ECC
Guides the agent to design and review REST APIs with consistent naming, status codes, pagination and versioning.
affaan-m/ECC
Write long-form articles and guides in a specific, non-generic voice with strong structure
affaan-m/ECC
Guides the agent to write and review Node.js, Express, and Next.js API code using standard backend patterns.
affaan-m/ECC
Scores competitors on nine weighted dimensions with explicit 1-5 rubrics and a tension plot
affaan-m/ECC
Guides a structured, multi-session brand identity interview and saves a resumable brandbook
affaan-m/ECC
Builds a reusable writing-style profile from your real posts and docs so drafts sound consistent.
affaan-m/ECC
Guides an agent through using Bun for JavaScript/TypeScript running, installs, bundling, testing, and Vercel setup
affaan-m/ECC
Maps what a planned code, data, or interface change will affect, then checks the finished diff against that plan.
affaan-m/ECC
Applies shared naming, readability, immutability and code-quality conventions when reviewing code.
affaan-m/ECC
Scopes and tiers a competitor set before any benchmarking, using a client positioning brief.
affaan-m/ECC
Assembles scored competitor cards into a structured, decision-ready competitive report
affaan-m/ECC
Turns source material into platform-native posts, threads, video scripts and campaigns without generic hype.
affaan-m/ECC
Keeps business term definitions in one CONTEXT.md and routes hard decisions to the ADR skill
affaan-m/ECC
Adapts one idea into distinct posts for X, LinkedIn, Threads and Bluesky without duplicating copy.
affaan-m/ECC
Runs multi-source web research and writes cited reports using firecrawl and exa tools
affaan-m/ECC
Run and coordinate several AI agent sessions in parallel using dmux and tmux panes
affaan-m/ECC
Routes docs-governance requests to the right ECC skill and runs an opt-in read-only docs audit.
affaan-m/ECC
Fetches current library and framework docs via Context7 instead of relying on memory.
affaan-m/ECC
Write stable, maintainable Playwright browser tests and fix flaky ones in CI
affaan-m/ECC
Teaches the agent ECC repo's commit, naming, and testing conventions
affaan-m/ECC
Lets an agent define, run and report pass/fail tests for Claude Code workflows before trusting or changing them.
affaan-m/ECC
Lets your agent search the web, code docs, companies and people via Exa's search engine.
affaan-m/ECC
Generate images, videos and audio with AI models via the fal.ai MCP server
affaan-m/ECC
Guides the agent to write and review React and Next.js components, state, and performance code
affaan-m/ECC
Builds animated, single-file HTML presentations that run in any browser
affaan-m/ECC
Drafts pitch decks, memos, financial models and fundraising docs that stay consistent with each other.
affaan-m/ECC
Draft concise, personalized emails to angels, VCs and accelerators with a clear ask
affaan-m/ECC
Run sourced market, competitor, and investor research that ends in a clear decision
affaan-m/ECC
Flags MCP config package references that could silently pull in newer, different code
affaan-m/ECC
Guides building Node/TypeScript MCP servers with tools, resources, prompts, and Zod validation
affaan-m/ECC
Guides an agent to build, review, and harden production ML systems with data contracts, evals, and rollback plans.
affaan-m/ECC
Sets up local agent rules and domain-doc links for a business module without duplicating project docs
affaan-m/ECC
Checks that changing one module doesn't silently break the modules that depend on it.
affaan-m/ECC
Guides an agent through Next.js 16+ dev speed, Turbopack vs webpack, and bundle optimization
affaan-m/ECC
Lets the agent open plans in a local browser review page where you annotate, chat, and approve or request changes
affaan-m/ECC
Turns product ideas and PRDs into an implementation-ready capability plan with constraints and open questions
affaan-m/ECC
Gives the agent a security checklist and secure-coding patterns to review code against.
affaan-m/ECC
Checks ECC skills for frontmatter and Claude-only substitutions that break in Codex or Cursor
affaan-m/ECC
Nudges you to run /compact at good task breaks instead of letting auto-compaction cut in mid-task
affaan-m/ECC
Makes the agent write tests before code and aim for 80%+ coverage
affaan-m/ECC
Track a project's tests, required coverage, and bug-regression evidence in one registry
affaan-m/ECC
Share saved context and handoffs between AI agents through a local, inspectable memory vault.
affaan-m/ECC
Runs build, type, lint, test, security and diff checks before a Claude Code session claims work is done
affaan-m/ECC
Guides an agent through cutting, structuring and polishing real video footage with FFmpeg, Remotion and AI tools.
affaan-m/ECC
Lets an agent post tweets and threads, read timelines, search X, and pull analytics via the X API.
affaan-m/ECC
Drafts long-form articles, guides and newsletters in a consistent, non-generic voice
affaan-m/ECC
Turns one idea or source asset into platform-native posts, scripts, and content calendars.
affaan-m/ECC
Drafts personalized cold emails, warm intros, follow-ups and investor updates for fundraising
affaan-m/ECC
Guides an AI agent through eval-first development, task splitting, and model tier selection
affaan-m/ECC
Guides Claude Code through running unattended, looped development pipelines and multi-agent workflows
affaan-m/ECC
Teaches an agent to cache costly file processing results keyed by SHA-256 content hash.
affaan-m/ECC
Makes the agent write and review C++ following the C++ Core Guidelines for safe, modern code.
affaan-m/ECC
Guides the agent through writing, fixing and debugging C++ tests with GoogleTest and CTest
affaan-m/ECC
Plan safe, reversible database schema changes with zero-downtime patterns
affaan-m/ECC
Guides an agent through CI/CD, Docker, health checks and rollout strategies for web apps
affaan-m/ECC
Guides an agent through Django project structure, DRF APIs, ORM, caching and settings best practices.
affaan-m/ECC
Teaches the agent Django security best practices for auth, permissions, CSRF, SQL injection and XSS.
affaan-m/ECC
Guides the agent through Docker and Docker Compose setup, networking, volumes and container hardening.
affaan-m/ECC
Guides an agent to write and review production FastAPI services with best-practice structure.
affaan-m/ECC
