Skip to content
SkillModel-led599 tokens in use

to-spec

Synthesize the current conversation into a structured spec and publish it to your issue tracker.

328,764 stars on the repository MIT Licensemattpocock/skills · skills/engineering/to-spec
claude plugin install mattpocock-skills@claude-plugins-official
View source

Install command as written in the repository's README

Skill label

measured from its files

What installing to-spec costs

31tokens idle
599tokens in use

Idle is what sits in context while the skill is installed. In use is its instructions, loaded when it runs. Counts are approximate, made offline with cl100k_base.

Model dependence Model-led

Script-ledMixedModel-led

Mostly instructions, so how well it works depends on the model that follows them.

This describes the skill's structure, how much of it is scripts. It does not say how well it works.

Touches

Nothing detected: no shell, network, file writes, git or credentials in its files.

Declared tags come from the skill's frontmatter; detected ones from its scripts and code blocks. This is detected, not audited.

How it scores

Quality

How well the skill is written, counted from its own files: its frontmatter, its steps, examples, stated output and constraints, its size and whether every file it ships is one it names.How we score it →
Fair
  • Frontmatter8/15
  • Structure23/25
  • Examples3/15
  • Output10/15
  • Size15/15
  • Resources0/15

Security

Starts at 100 and loses points for each thing in the skill's files that a reader should look at first.How we score it →
Excellent

No findings.

Repository health

Pre-computed score based on development activity, maintenance, community, maturity, and trend momentum.How we score it →
Excellent
  • Development activity96
  • Maintenance100
  • Community56
  • Maturity32
  • Momentum40

to-spec takes what you and the agent have already discussed, plus its understanding of the codebase, and writes a spec from a fixed template. It does not interview you. It sketches the test seams for the feature, checks them with you, publishes the spec to the project issue tracker and applies the ready-for-agent label.

It expects the tracker setup from setup-matt-pocock-skills.

What You Get

  • A spec issue on your tracker with seven fixed sections
  • A long list of user stories in the as-an-actor format
  • Implementation decisions written without file paths or code snippets that go stale
  • A proposed set of test seams, ideally one

Common Use Cases

  • Capture the result of a long design chat as a ticket an agent can pick up
  • Record the modules and interfaces a feature will touch before work starts
  • Hand a feature to a background agent with testing decisions attached

What Makes It Unique

It synthesizes and never interviews, which makes it a fast closing step after a grilling session. It prefers the highest existing test seam and aims for as few seams as possible, and it applies the ready-for-agent label on publish so no further triage is needed.

Under The Hood

A single SKILL.md with an embedded spec template and no scripts. It explores the repo, uses the domain glossary vocabulary, respects ADRs, then publishes to whichever tracker setup-matt-pocock-skills configured, which can mean running gh or writing local markdown files.

Join founders building with open source

Opinionated takes, migration guides, cost-saving tips, and insights from the open source ecosystem.

Subscribe on SubstackJoin 750+ subscribers