mattpocock-skills
claudeMatt Pocock's agent skills for real engineering (grilling, spec/ticket flows, TDD, code review, domain modelling and more).
39 skills
Eleven tested Claude Code subagents with least-privilege tools and explicit output contracts: code review, security audit, test writing, debugging, flaky-test hunting, git archaeology, doc sync, PR writing, dependency audit, CLAUDE.md authoring, migration review.
Plugin subagents from marketplace agent37-subagents in agent37-platform/claude-code-subagents. Commands read from the repo's manifest — 1 month ago.
claude-md-writerWrites or updates a CLAUDE.md that actually helps agents work in this repo. Verified commands, conventions, gotchas, map of the code. Use on a repo missing one, or when the existing file has drifted from reality.
code-reviewerReviews the current diff for correctness bugs before you commit or open a PR. Use proactively after completing a feature or fix. Reports severity-ranked findings with file:line evidence; never edits files.
debuggerRoot-causes a failing test, error, or wrong output through hypothesis-driven investigation. Use when something fails and the cause is not obvious from the error alone. Returns the root cause with evidence plus the minimal fix.
dependency-auditorVets new or upgraded dependencies before they land. Advisories, maintenance health, licenses, install scripts, typosquats, transitive baggage. Use whenever a diff touches a package manifest or lockfile.
doc-syncerFinds and fixes documentation drift after a code change. README claims, examples, comments, and CHANGELOG entries that no longer match the code. Use after a feature lands or before a release.
flaky-test-hunterDiagnoses a flaky test by making it fail reliably, then fixes the actual nondeterminism. Use when a test passes locally but fails in CI, or fails intermittently. Reports the flake class with reproduction evidence.
git-archaeologistExplains why code is the way it is by tracing it through git history, blame, pickaxe searches, and linked PRs. Use before changing code that looks wrong or deleting code that looks dead. Read-only.
migration-reviewerReviews database schema migrations for destructive operations, lock hazards, deploy-order traps, and irreversibility before they run against shared data. Use on any diff containing migration files or DDL. Read-only.
pr-writerDrafts the pull request title and body from the branch's actual commits and diff, following the repo's PR template and conventions. Use when work is committed and ready to open.
security-auditorAudits changed code for exploitable vulnerabilities (injection, authz gaps, secrets, path traversal, SSRF). Use before merging anything that touches input handling, auth, file paths, subprocess calls, or HTTP. Read-only.
test-writerWrites tests for the code you just changed, in the repo's existing test idiom, and proves each test fails without the change. Use after implementing a feature or fix that lacks coverage.
Matt Pocock's agent skills for real engineering (grilling, spec/ticket flows, TDD, code review, domain modelling and more).
39 skills
Persistent memory for Claude Code: the next session starts where the last one left off.
29 skills · 1 MCP server · hooks
Open-source AI job search that runs locally in your AI coding CLI: scan job portals, evaluate listings into a structured A-H report with a 1-5 score, tailor your CV, track applications
5 skills · hooks
Cross-session memory and token savings for coding agents.
6 skills · 1 agent · 1 MCP server · hooks
Up-to-date documentation lookup. Pull version-specific documentation and code examples directly from source repositories into your coding agent's context.
1 MCP server
Automates browser interactions for web testing, form filling, screenshots, and data extraction
12 skills
This page updates itself from your repo. Add the badge to your README, or get it Featured to pin it above the organic results with a followed link to your site.