specify
Skill4 starsChanged 57 days agoApache-2.0View on GitHub
Turn an idea into a design document a decomposition can read. Use for /spec-ify, /specify, or when an idea cannot yet produce issues because something has to be decided first.
29 other projects use a version of this.
Adopted by
Most recently taken up by
- ZaxbyHub/opencode-swarmtheir own version, first seen Nov 15, 2026
- danielcherubini/skillstheir own version, first seen Nov 11, 2026
- joeldevz/skynex-pitheir own version, first seen Nov 8, 2026
- Buddhilive/buddhi-clitheir own version, first seen Nov 1, 2026
- jantichy/claudetheir own version, first seen Oct 27, 2026
- totalfrank/skillstheir own version, first seen Oct 27, 2026
Used with
Projects with it also use discovery, discover and 10 more.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
robzolkosSkilldiscovery
Quick user-centric interview to capture requirements from a time-poor stakeholder.
Also in 3 projects that use this
anombyte93Skilldiscover
Phase 1 of the prd-taskmaster pipeline: brainstorm-driven discovery. Delegates to superpowers:brainstorming in Interactive Mode (one adaptive question at a time), or self-brainstorms in Autonomous Mode when no user is present. Intercepts before the brainstorming chain hands off to writing-plans — this skill owns the exit. Extracts constraints, calibrates scale (Solo / Team / Enterprise), and advances the pipeline to GENERATE.
Also in 4 projects that use this
MicrosoftSkillspec-review
Independent, multi-model review of a design or spec **before** any code is written, for the microsoft/winappcli repo. Activate when a contributor asks to "review this spec", "review my design", "review this design doc", "validate this approach", "should we build this", "spec review", "design review", or "feature review". Fans out parallel sub-agents — each doing its OWN research against the real codebase and ecosystem rather than trusting the spec — covering necessity & scope, approach & alternatives, feasibility vs reality, risks & unknowns, DX & user impact, and a different-model-family cross-check. Emits a decision-oriented recommendation (proceed / proceed-with-changes / reconsider) to stdout. This is the PRE-CODE companion to the pr-review skill (which reviews code already written); use spec-review at the design/spec stage, not on an implemented diff. Does NOT write code or edit the spec.
Also in 3 projects that use this
MicrosoftSkillinit
Also in 5 projects that use this
MetaSkillroadmap
Build, continue, inspect, or visualize a source-grounded mathematical roadmap and theorem DAG in an existing Autoform Markdown blueprint. Use for source research, repository-wide inventories of existing and planned mathematics, scope, coverage, milestones, and roadmap articles; do not create repository infrastructure or prove Lean declarations.
Also in 3 projects that use this
sahinSkillexecute
Also in 3 projects that use this
Other versions29 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 29
- ZaxbyHub/opencode-swarm★ 531same name
- zaxbysauce/opencode-swarm★ 519same name
- genkovich/sdd★ 187same name
- team-attention/hoyeon★ 173same name
- ArvorCo/Relentless★ 98same name
- jantichy/claude★ 52same name
- mehrad-dm/mastermind★ 23same name
- agent-ix/quoin★ 20same name
- danielcherubini/skills★ 14same name
- totalfrank/skills★ 4same name
19 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository14 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- commits-and-prsSkill
- decomposeSkill
- github-boardSkill
- implementSkill
- releaseSkill
- reviewSkill
- reviseSkill
- rubber-duckSkill
- thunderstormSkill
- triageSkill
- worktreeSkill
- writing-docsSkill
The file
We split this skill into the rules it gives an agent
4 rules. 0 are common, 3 are distinctive, and 1 is very specific to this project.
0
Common
Most of the 27 edited versions we read keep these.
3
Distinctive
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 14
Ask one question of the idea: can you write a sub-issue's stop rule without deciding anything else first?
- Very specific: line 17
If yes, there is nothing for a spec to add.
- Distinctive: line 21
If no, the missing decision is what the spec is for.
- Distinctive: line 91
Write a spec for work whose criteria you could already state.
“Common” is measured: kept by at least 6 in 10 of the 27 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Specify
One design, written down, so that work can be cut from it. A spec exists to make a decision that issues would otherwise each have to make separately and inconsistently.
Write one only when it earns itself
Ask one question of the idea: can you write a sub-issue's stop rule without deciding anything else first?
If yes, there is nothing for a spec to add. Go to decompose and file the
work. Most chores, bug fixes, and anything already described by a finding are
in this case.
If no, the missing decision is what the spec is for. "Give tool activities distinct semantic kinds" cannot produce a checkable stop rule until someone settles what the kinds are, and five issues that each invent their own answer is worse than one document that picks.
A spec that records no decision is a summary of the idea with more words.
Where it goes
internal/features/<feature-name>/plan.mdOne directory per feature track, named for the feature rather than for the
change. Every file under internal/ carries the frontmatter that
internal/thunderstorm.md requires:
---
name: <short-kebab-case-name>
last_updated: <YYYY-MM-DD>
id: <ULID>
---Generate the identifier with .claude/scripts/ulid.py. The idea the spec came
from cites its own identifier, and issues cut from the spec cite the spec's, so
the trail reads from either end.
What a spec holds
- The decision. What was chosen, and what it rules out. This is the reason the document exists.
- The shape. How the pieces fit: boundaries, ownership, the vocabulary the issues will use. Enough that two people cutting issues from it cut the same ones.
- What it does not cover. Scope the work must not absorb, so a worker who finds adjacent work knows it is adjacent.
What a spec does not hold
No task list. The board holds the work. A checklist inside a spec is a
second copy of the board's state that nothing updates and nobody reviews, and
it goes stale the first time an issue closes. decompose reads the spec and
files issues; that is what replaces the checklist.
The older tracks under internal/features/ still carry a tasks.md beside
their plan. Those predate the board and are not a pattern to copy. Do not add
one.
No implementation. Code in a spec is a sketch nobody compiles. Name the types and the boundaries if they carry the decision; stop there.
Review
A spec is a document with a separate review, which is the whole reason the
rubber-duck skill refuses to expand an idea into one. Open it as a pull
request and run /rev against that, the same as any other change. A spec that
merges unreviewed sets the vocabulary for every issue cut from it.
Then
Hand it to decompose once the pull request above has merged. Issues cite the
spec's identifier and a worker reads the spec from a worktree cut from edge,
so issues filed off the branch point at a file nobody else has, and at decisions
review can still change. A spec that produces no issues is either premature or
was not needed.
Do not
- Write a spec for work whose criteria you could already state.
- Restate the idea. Link it and record what the idea left open.
- Add a
tasks.md. - File issues here. That is
decompose.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

