mattpocock-skills
claudeMatt Pocock's agent skills for real engineering (grilling, spec/ticket flows, TDD, code review, domain modelling and more).
39 skills
Helps legalquants/lq-ai maintainers triage PRs, issues, and dependabot traffic against lq-ai's canon. Recommends outcomes and drafts everything onto one reading deck — the warm public comment, the merge message, one-click suggestion fixes — and records what the maintainer decided. A human decides, every time.
Plugin lq-maintainer from marketplace lq-maintainer-agent in houfu/lq-maintainer-agent. Commands read from the repo's manifest — 3 days ago.
This plugin runs code on your machine — its hooks fire automatically on events like tool use. The exact commands are listed below; read them before installing.
design-planThe design path for category-1 contributions — greenfield / new-feature work (the DE series), on a PR or an issue. Produces a plan instead of a code review: a warm acknowledgment, the decision inventory with a drafted ADR or DE stub per open decision, the predicted obstacles, and the ordered decompo
labelThe express first touch: label one open item (or sweep the open queue) from a PROVISIONAL classification made with the same rule files the router uses, projected onto the target repository's own existing label taxonomy (rules/labels.md LB-NN). No deck, no receipt, no digest, no findings, no tier — s
milestoneScan one milestone of the target repository: what is LEFT in it and what is blocking it, over the OPEN PRs and issues carrying that milestone, in four buckets (blocking / ready to close out / needs work / not yet assessed). Reads the evidence the fuller passes already recorded rather than re-reviewi
release-notesDraft the release narrative for the target project over a commit range (default last-tag..HEAD): breaking changes first with their migration and undo lines, then features, fixes, and the rest by change category, with contributor credit per item, a drafted-never-decided semver suggestion, and a prove
review-issueConsidered review of a single issue in the target project — the issue counterpart to review-pr. Classifies, performs its own cross-reference (never the filer's), previews the obstacles and tier the issue would hit as a PR, and produces the recommendation deck (needs-info / decompose / proceed / desi
review-prTiered review of a single pull request in the target project. The default is the Tier-1 quick pass — one time-boxed, single-context read of the diff that ends in exactly one concrete action (merge / merge-after-one-named-fix / discuss-a-specific-question / route-to-design) with its undo path stated.
triageTriage the target project's open queue: the batch router that sorts every open PR and issue into recommended lanes, classifies each into a change category (G-NN) and, for behavioral-change/bug-fix items, a review tier (TR-NN), and draws an action-first digest (outcome + tier + undo path leading, lan
review-passDispatch-only worker for /lq-maintainer:review-pr Step 3 — one member of the four-pass review team (anchor/scope, security-vetting, code-quality, test-adequacy). Use this agent ONLY when the review-pr skill's Step 3 dispatches a pass; it is never invoked proactively, never for general tasks, and nev
triage-itemDispatch-only worker for /lq-maintainer:triage batch mode — the per-item router fan-out (design §3.3). One of these judges exactly one open PR or issue: lane, change category, review tier, and the item's own outputs, in its own fresh context. Use this agent ONLY when the triage skill fans out in bat
sh "${CLAUDE_PLUGIN_ROOT}/settings/hooks/block-writes.sh"
Matt Pocock's agent skills for real engineering (grilling, spec/ticket flows, TDD, code review, domain modelling and more).
39 skills
Core skills library for Claude Code: TDD, debugging, collaboration patterns, and proven techniques
15 skills · hooks
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
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.