mattpocock-skills
claudeMatt Pocock's agent skills for real engineering (grilling, spec/ticket flows, TDD, code review, domain modelling and more).
39 skills
Spec-Driven → Test-Driven development pipeline with specialized agents
Plugin sdd from marketplace claude-sdd in Francisco-Donadio/claude-sdd. Commands read from the repo's manifest — 12 days ago.
publish-stackPush a stacked-PR plan built by /sdd:run and open its PRs bottom-up, each based on the previous slice's branch.
runRun the full SDD→TDD agent pipeline for a feature, delegating to the specialized agents.
architectDefines the technical architecture — which module/store/service owns the change, data flow, integration points, and the trade-offs behind those choices. Use after the spec is approved, before detailed design and coding.
archivistCloses out a completed feature. Summarizes everything that was done into a single durable archive record and tidies the feature folder. Use LAST, after verification passes.
designerDetailed design — concrete interfaces, function/endpoint signatures, types, data schemas, and module boundaries derived from the architecture. Use after architecture, before tests are written (tests are written against this design).
explorerRead-only codebase reconnaissance. Use FIRST in the pipeline (or anytime you need to understand existing code) to map the files, modules, conventions, and integration points relevant to a request. Never modifies code.
implementerWrites the production code to make the failing tests pass (TDD green step), following the design and tasks. Use after the test-author. Implements the minimum needed to satisfy tests and spec — no scope creep.
reviewerLightweight code review of the implementer's diff — ONE focused pass for correctness bugs, spec/design deviations, and convention violations. The pipeline's cheap default review tier (a single agent, no fan-out); the heavier `/code-review` skill covers the standard/deep tiers. Read-only — reports fi
spec-writerSpec-Driven Development author. Turns a feature request plus exploration findings into a clear, testable specification (the WHAT and WHY, not the HOW). Use after the explorer and before planning/architecture.
task-plannerBreaks an approved spec into an ordered, dependency-aware list of small, verifiable tasks. Use after the spec is approved and before/alongside architecture so work can be tracked and sequenced.
test-authorTest-Driven Development author. Writes failing tests FIRST against the design's contracts, before any implementation exists. Use after design and before the implementer. Confirms the tests fail for the right reason (red phase).
verifierIndependent verification. Runs the full test suite, lint, and checks the implementation against the spec's acceptance criteria. Use after the implementer. Reports PASS/FAIL with evidence — does not fix code itself.
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
5 skills + 35 modes + Material Passport pipeline. Includes v3.6.7 cross-model audit gate and v3.6.8 generator-evaluator contract.
5 skills · 16 commands · 4 agents · 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.