Matt Pocock Agent Workflow Suite - Setup Guide
Source: mattpocock/skills (20 skills, 11.4M+ combined installs)
Category: Software Engineering / Agent Workflow / Product Management
First Seen: Sep 9, 2026
This guide covers 20 Matt Pocock skills that were not included in the earlier Mattpocock Skills Setup (19 skills) or Engineering Skills (5 skills) guides. Together they form the interrogation-and-execution layer of the Pocock methodology: grilling plans, TDD loops, domain modeling, triage, handoffs between agent sessions, and spec-to-ticket pipelines. Combined installs: 11,400,000+.
Installation
Install all 20 at once:
# Interrogation layer
npx skills add mattpocock/skills --skill grill-me
npx skills add mattpocock/skills --skill grill-with-docs
npx skills add mattpocock/skills --skill loop-me
npx skills add mattpocock/skills --skill wait-what
# Engineering layer
npx skills add mattpocock/skills --skill tdd
npx skills add mattpocock/skills --skill code-review
npx skills add mattpocock/skills --skill diagnosing-bugs
npx skills add mattpocock/skills --skill domain-modeling
npx skills add mattpocock/skills --skill research
npx skills add mattpocock/skills --skill setup-ts-deep-modules
# Planning-to-tickets layer
npx skills add mattpocock/skills --skill to-spec
npx skills add mattpocock/skills --skill to-tickets
npx skills add mattpocock/skills --skill to-questionnaire
npx skills add mattpocock/skills --skill wayfinder
npx skills add mattpocock/skills --skill wizard
# Session and collaboration layer
npx skills add mattpocock/skills --skill handoff
npx skills add mattpocock/skills --skill triage
npx skills add mattpocock/skills --skill teach
npx skills add mattpocock/skills --skill writing-for-agents
npx skills add mattpocock/skills --skill retro
Prerequisites
| Requirement |
Details |
| Hermes Agent |
Any recent version with skill support |
| Node.js 18+ |
For the npx skills CLI |
| GitHub repo with Issues enabled |
For triage, to-spec, to-tickets, wayfinder output |
GitHub PAT with repo scope |
For issue/ticket creation |
Capabilities
Interrogation Layer
| Skill |
Installs |
What It Does |
| grill-me |
1,103,281 |
A relentless interview to sharpen a plan or design. Dispatches to grilling. The flagship skill of the suite |
| grill-with-docs |
941,573 |
The grilling interview that also creates docs (ADRs and glossary) as it goes |
| loop-me |
277,934 |
Grill me about specs for the workflows I want to build, within this workspace |
| wait-what |
224,639 |
"Stop. That last message did not land: re-pitch it." - clarity catch |
Engineering Layer
| Skill |
Installs |
What It Does |
| tdd |
873,338 |
Red-green-refactor loop reference: what a good test is, where tests go, anti-patterns, loop rules. Reads CONTEXT.md and ADRs |
| code-review |
521,103 |
Reviews changes since a fixed point along two axes: Standards (repo's documented coding standards) and Spec (does the code match what was asked) |
| diagnosing-bugs |
574,167 |
Diagnosis loop for hard bugs and performance regressions |
| domain-modeling |
603,161 |
Build and sharpen a project's domain model; writes/edits CONTEXT.md and ADRs |
| research |
476,454 |
Investigates a question against high-trust primary sources, captures findings as a Markdown file in the repo |
| setup-ts-deep-modules |
211,966 |
Wires dependency-cruiser into a TypeScript repo so each package is a deep module (implementation hidden, reachable only through entry points) |
Planning-to-Tickets Layer
| Skill |
Installs |
What It Does |
| to-spec |
474,307 |
Turns the current conversation into a spec and publishes it to the issue tracker - no interview, just synthesis |
| to-tickets |
466,073 |
Breaks a plan/spec/conversation into tracer-bullet tickets, each declaring blocking edges, published to the configured tracker |
| to-questionnaire |
282,849 |
Turns a decision you can't fully answer into a questionnaire for someone else |
| wayfinder |
483,129 |
Plans work too big for one agent session as a shared map of decision tickets, resolved one at a time |
| wizard |
335,511 |
Generates an interactive bash wizard for steps only a human can perform (infrastructure, credentials, CI secrets, third-party dashboards) |
Session and Collaboration Layer
| Skill |
Installs |
What It Does |
| handoff |
774,587 |
Compacts the current conversation into a handoff document for another agent to pick up |
| triage |
760,099 |
Moves issues and external PRs through a state machine of triage roles: categorise, verify, grill if needed, write agent-ready briefs |
| teach |
626,663 |
Teaches the user a new skill or concept, within this workspace |
| writing-for-agents |
233,014 |
Writing documents for agents: creating/editing skills, AGENTS.md, CLAUDE.md |
| retro |
72,735 |
Conducts a retrospective on a coding session |
Quick Start
- Install the interrogation layer first:
npx skills add mattpocock/skills --skill grill-me
- Grill a plan before building: invoke
grill-me (or grill-with-docs if you want ADRs and a glossary as output)
- Build with
tdd, review with code-review, diagnose failures with diagnosing-bugs
- Convert the finished conversation to specs and tickets:
to-spec then to-tickets (or wayfinder for multi-session work)
- End each session with
handoff so the next agent session picks up without re-reading the transcript
Typical workflow: grill-me -> tdd -> code-review -> to-spec -> to-tickets -> handoff
Limitations / Verification
# Verify skills installed
hermes skills list | grep -E "grill-me|tdd|handoff|triage|to-spec"
# Functional test
# "Grill me about this plan" -> invokes grill-me -> dispatches to grilling
grill-me sets disable-model-invocation: true - it is invoked via the Skill tool, not by the model directly
loop-me, retro, setup-ts-deep-modules live in the repo's in-progress directory - stable but evolving
- Marketplace-only siblings (listed on skills.sh, no SKILL.md in the repo main branch yet):
qa (205K), diagnose (238K), obsidian-vault (201K), review (94K), batch-grill-me (64K)
Security