Dev & Engineering

Ask Matt Skill Router

Recommends the best skill or workflow for your engineering situation, so you don't have to memorize every skill.

50/ 100
Use with care

Useful, but reliability, evidence or controls still have material gaps.

See how it was scored ↓
Works as-is in
Codex · Claude Code
Stars
★ 281k
Last updated
3d ago
License
MIT
skill-routerworkflow-orchestrationengineering-workflowscode-quality

What does this skill do, and when should you use it?

Ask Matt is a routing skill that helps developers choose the most appropriate skill or workflow from a complex skill set. It details a complete path from idea to shipping, including requirements elicitation, prototyping, specification, ticket breakdown, implementation, and code review, along with standalone skills and context management strategies. Created by well-known developer Matt Pocock, it serves as an entry point to the engineering skills in this repository.

This skill takes a description of your engineering task and matches it to the appropriate skill or flow based on context (e.g., working directory presence, task size, multi-session needs). It reads the repository structure and skill definitions, then provides a recommendation with explanation of the workflow path. It does not execute engineering tasks itself but acts as a navigation and routing aid.

Good fit
  • A developer new to this skill set doesn't know which skill to start with for feature development.
  • A user has a new idea and needs to refine it through an interview-style discussion.
  • A user faces a large, vague engineering effort requiring multi-session planning.
  • A user has a backlog of bugs and requests that need triaging first.
  • A user needs to decide whether to continue, clear, hand off, or compact context at a phase boundary.

How do you install this skill?

Before you use it
  • This skill references other skills; its actual effectiveness depends on those skills. Users should verify the safety and reliability of the dependent skills.
  • Data flow and permission details are not disclosed; if dependent skills involve sensitive data or external operations, additional review is needed.
  • Mainland China network reachability is not specified, which may affect usage of skills dependent on overseas services.
Before you start
Your agent needs
  • Local filesystem

Install via npx skills@latest add mattpocock/skills, or as a Claude Code plugin. Ensure you select this skill (ask-matt) and /setup-matt-pocock-skills, then run /setup-matt-pocock-skills before first use.

Generic route: install into Claude Code manually (macOS / Linux)
tmp="$(mktemp -d)"
git clone --depth 1 https://github.com/mattpocock/skills.git "$tmp"
mkdir -p ~/.claude/skills
cp -R "$tmp/skills/engineering/ask-matt" ~/.claude/skills/
rm -rf "$tmp"

Generated from the source repository and skill path; it copies only this skill's folder. If the author's install steps above differ, follow those first. To scope it to one project, replace ~/.claude/skills with that project's .claude/skills.

How do you use this skill?

Try saying

Once installed, send your agent any of these to trigger it:

  • I have a new idea and want to start building.

Invoke /ask-matt in a compatible coding agent and describe your situation, e.g., 'I have a new idea and want to start building.' The skill will guide you to the next step.

What are this skill's strengths and limitations?

Pros
  • Provides a clear workflow map covering the entire path from idea to shipping.
  • Supports diverse scenarios (multi-session, prototyping, triage, architecture improvement).
  • Emphasizes context hygiene to avoid model degradation.
  • Open-source and freely modifiable for personal adaptation.
Limitations
  • Highly dependent on other skills in the repo; limited value in isolation.
  • Complexity may overwhelm beginners.
  • Requires pre-configuration of tools like issue trackers.
  • No automated tests or release process for the skill itself.

How does this skill compare with similar options?

Side by side with related skills; every score comes from the same FSRS standard.

Skill FS score Stars Last updated License
Ask Matt Skill Router this page 50 · Use with care ★ 281k 3d ago MIT
Expo Skill Eval 48 · Use with care ★ 2.7k 6d ago MIT
Verification Before Completion 48 · Use with care ★ 297k 16d ago MIT
Engineering Workflow Navigator 46 · Use with care ★ 103k 8d ago MIT
Superpowers Skill Bootstrap 35 · Not recommended ★ 297k 16d ago MIT

This skill is the entry router for the repository, not an alternative to other skills. It has no direct comparison, as it is part of a larger skill set.

How did FollowSkills review this skill?

FollowSkills review · FSRS-2.0
Use with care
50/ 100 5-point scale 2.5 / 5
The upstream repository has new commits since this review. The score still applies to the reviewed revision shown and may not cover the latest changes.
1Trust17 / 25 · 3.4/5

The skill is a router/documentation type, does not itself execute code, and has low permission requirements. However, it references other skills in the repo, which may involve external effects and sensitive data operations, and the current file does not clearly specify data flow, confirmation mechanisms, or rollback strategies. Deductions accordingly.

2Reliability6 / 20 · 1.5/5

The skill mainly depends on other skills; its own instructions are mostly routing logic. Static review cannot verify its execution results, and there are no tests or failure feedback mechanisms. Hence low score.

3Adaptability8 / 15 · 2.7/5

The skill targets Matt Pocock's engineering workflow; audience is clear, but boundaries and non-fit scenarios are not explicitly declared, trigger conditions depend on prerequisite skills, and mainland China network reachability is not considered.

4Convention11 / 15 · 3.7/5

Documentation structure is clear with phase breakdowns, but lacks installation instructions, version history, maintenance responsibility, and clear disclosure of known limitations.

5Effectiveness5 / 15 · 1.7/5

As a router, its value lies in guiding users, but actual effectiveness is not verified; output depends on other skills' performance, and static review cannot confirm direct usability.

6Verifiability3 / 10 · 1.5/5

Only documentation and example references provided; no tests or other evidence, conclusions cannot be independently reproduced.

1 2 3 4 5 6

Open a dimension to read why it scored that way

Reviewed Aug 07, 2026 Reviewed revision 84fdeffd12f2 Review evidence[1][2][3][4][5][6][7]

Evidence confidence:Low — Mostly static review, author material or a limited demo; useful for discovery, not high-risk decisions.

See the full review method →

FAQ

Is this skill free to use?
Yes, it's open-source under the MIT license. However, running it requires a compatible AI model and possibly a subscription to tools.
Does it work with all coding agents?
It follows the Agent Skills standard, so it should work with any agent that supports it, but some features like disable-model-invocation may be Claude Code-specific.
What if I don't like its recommendation?
It's advisory; you can override and invoke other skills directly, or modify the skill to suit your workflow.

More skills from this repository

All from mattpocock/skills

Related skills