mattpocock-skills
claudeMatt Pocock's agent skills for real engineering (grilling, spec/ticket flows, TDD, code review, domain modelling and more).
39 skills
Ace in my space -- a suite of agent skills for scoping, delivering, and reviewing work in a workstream, with pull requests, cross-repository consultation, report publication, and harness detection.
Plugin ace from marketplace ace-in-my-space in ardallie/ace-in-my-space. Commands read from the repo's manifest — 4 days ago.
agent-consultantAnswers a cross-repository consultation request as the consultant side of the sibling-exchange protocol -- spawns one to four researchers plus a sceptic to research this repository's capabilities, contracts, and intent, and produces a self-contained answers report the user carries back to the reques
detect-harnessPrints the Codex agent model and reasoning-effort mapping for the suite's three capability tiers, read from models.md beside this skill. Invoke before spawning tiered agents when no valid cached resolution is present; see the cache check for reuse and invalidation rules.
git-branchRewrites the current feature branch for one purpose -- pr (logical commits) or rebase (onto the latest default branch) -- then force-pushes it under a pinned lease, keeping a local backup. Runs only on an explicit request to rewrite the current branch.
pr-createCreates a GitHub pull request from current branch commits: analyses all commits since branching and generates a PR description following the repository template.
pr-readReads one or more pull requests (bare number or full URL) into the session: metadata, description, and all three comment surfaces (conversation comments, inline review comments, review bodies); --comments no skips the comment surfaces; read-only.
pr-responseResponds to PR review comments based on work done in the conversation: reads the PR via $ace:pr-read, gates on the fixes being pushed to the PR head branch, then posts a summary comment and threaded replies.
report-publishPublishes any report markdown file (scope envelope, scope review, scope code review, consultation answers) as a new GitHub issue; the title composes from the artefact's Title:/Type:/Stage: header fields with the H1 as fallback; path argument is required.
run-interviewPresents open questions interactively through the Codex question surface; every option list includes keep-current behaviour and marks exactly one option Recommended.
run-retroWrites a retrospective report on a finished run to .ace/retro/ -- a run is any unit of work completed in this session: a slash command or skill invocation, a multi-agent orchestration, or an ad-hoc task. The report exists to improve the process that ran -- the command or skill instructions, sequenci
scope-code-reviewReviews a workstream's delivery at code level before its pull request -- correctness, robustness, security, tests, maintainability, stage seams and skill text -- reading the scope and the decision log as settled instead of re-arguing them, reading the delivery records for what was delivered, and end
scope-envelopeOpens a workstream by defining the scope of a large or loosely specified ambition: a knowledge-first multi-agent deliberation, sceptic always included, producing a scope envelope -- direction, goals, assumptions and suggested stages with observable gates -- beside the workstream's decision log. $ace
scope-implementDelivers a workstream's current scope report -- implements its scope envelope, or remediates its closing review's stages -- carrying every stage through design, implementation, integration and verification as a multi-agent run that commits and pushes a feature branch and leaves a handoff beside the
scope-report-triageRemediates a workstream's current code review, or a pull request's review tied to its delivery, in a scope-report-triage run in that workstream: verifies each finding against HEAD, reads the scope and the decision log as settled, takes any change to the scope, a goal, a gate clause or an accepted de
scope-reviewCloses a workstream by reviewing delivered work against its scope -- envelope and decision log, or scope documents: did goals land, were decisions honoured, was anything silently deferred, what remediation follows. Multi-agent and architectural, sceptic included. Not a review of the scope definition
ws-createFinds or creates a workstream -- a .ace/ws/{yyyyMMdd}-{slug}/ directory, a new one holding index.md, inputs.md, decisions.md, prompts/ and workings/ -- for given inputs or a name, reusing the one those inputs already belong to, and prints its path.
Matt Pocock's agent skills for real engineering (grilling, spec/ticket flows, TDD, code review, domain modelling and more).
39 skills
Talk like caveman. Cut filler. Keep technical details explicit.
22 skills · 1 command · 3 agents · hooks
Persistent memory for Claude Code: the next session starts where the last one left off.
29 skills · 1 MCP server · hooks
tasteskill 2.5.1: professional company websites at the figma.com and x.ai level, with one section per screen, real photographs, real product UI, effects kept for tech and gaming pages, and a real-browser checker and finish gate.
1 skill
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
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.