mattpocock-skills
claudeMatt Pocock's agent skills for real engineering (grilling, spec/ticket flows, TDD, code review, domain modelling and more).
39 skills
Autonomous feature-development pipeline for Claude Code: investigation → decisions → plan → review → implementation → codereview → report. Makes every decision itself (no interactive questions), with persistent state and a Stop-hook workflow gate.
Plugin ai-dp from marketplace claude-ai-dev-pipeline-plugin in kostia-official/claude-ai-dev-pipeline-plugin. Commands read from the repo's manifest — 1 month 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.
codereviewUse when an active ai-dp run is at the codereview step. Runs an embedded simplify review (reuse / quality / efficiency) over this run's diff and applies fixes. Self-contained — no external /simplify dependency.
decisionsUse when an active ai-dp run is at the decisions step. Autonomously resolves every hard design decision before planning and writes a decisions.md listing each decision, the options considered, and which one was selected (with rationale).
implementationUse when an active dev-pipeline run is at the implementation step. Generates todos from plan.md, implements them one-by-one, and runs typecheck + lint at the end. Strict-typing rules (no `any`, no dirty `as`) are self-policed by the skill and verified by the project's lint.
improveUse when the user asks to improve, fix, tweak, or extend the dev-pipeline plugin itself (its skills, orchestrator, scripts, manifest). Edits this plugin's source tree, bumps version, and creates a git commit. Plugin authors only — refuses to run if the plugin is installed via marketplace cache.
investigationUse when an active dev-pipeline run is at the investigation step. Gathers wide context about the feature and writes a structured context.md (Feature explanation, Related files, Existing code worth reusing, Risks & unknowns).
plan-improve-applyUse when an active ai-dp run is at the plan-improve-apply step. Walks through review.md issues one-by-one as todos and patches plan.md, deciding every unclear fix itself. Autonomous — never asks the user. Self-contained — embeds the original /plan-improve-apply discipline.
plan-improveUse when an active dev-pipeline run is at the plan-improve step. Reviews plan.md for correctness, missing pieces, and weak points; writes a numbered issue list to review.md. Self-contained — embeds the original /plan-improve discipline plus dp-specific reuse/extraction checks.
plan-wrapupUse when an active ai-dp run is at the plan-wrapup step. Final consistency pass on plan.md (and context.md), summarizes what changed during review, then auto-advances to implementation. Autonomous — no user sign-off.
planUse when an active ai-dp run is at the plan step. Writes a detailed plan.md that opens with a Proposal section (what/why/root cause/brief approach) and continues with Context, File-by-file changes, Reuse & extraction, and Verification. Reads decisions.md and honors it. Autonomous — never asks the us
reportUse when an active ai-dp run is at the report step (the final step, after codereview). Writes a readable report.md summarizing how the feature was implemented, problems encountered, any tech debt added, and what tests were added.
/ai-dp:dev-pipelineStart, resume, or continue an autonomous ai-dp feature-development run. Accepts a feature description (new run), a path to a state.json/context.md/plan.md (explicit resume), or a phrase like "continue <feature>" / "find plan for <feature>" (implicit resume).
bun ${CLAUDE_PLUGIN_ROOT}/scripts/hooks/capture-session.ts
bun ${CLAUDE_PLUGIN_ROOT}/scripts/hooks/enforce-pipeline-progress.ts
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
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
Build powerful, stateful CLI interfaces for any GUI application using the cli-anything harness methodology.
5 commands
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.