gap-check
Skill7 starsChanged 25 days agoMITView on GitHub
Silence detection for specs, plans, and ticket sets. Asks what is expected but absent, such as acceptance criteria, error paths, migrations with rollback, non-functional constraints, test anchors, and scope boundaries. Use on any adopted artifact and on every drafted spec before the human gate.
1 other project uses a version of this.
Adopted by
Most recently taken up by
- vatsal2210/claude-skillstheir own version, first seen Oct 15, 2026
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 1
- vatsal2210/claude-skills★ 0same name
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 repository2 other files it gives its agents.
- record-contractSkill
- spineSkill
The file
We split this skill into the rules it gives an agent
2 rules. About 0 are common, 2 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 23
Data and migration: If any stored shape changes, is the migration named, and is the rollback named?
- Distinctive: line 41
Never silently fill a gap.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Gap check
Reviewers check what is present. This skill checks what is missing. Run it on
intent.md, spec.md, plan.md, and tickets.json, whether Spine produced
them or /spine:spine adopt imported them.
The six absence questions
Answer each with present, absent, or partial, with a line reference for
present and partial.
- Acceptance criteria. Are they present, numbered, and testable? A criterion is testable when a reader can name the input, the action, and the observable result.
- Error paths. For every action the artifact names, is the failure case named: invalid input, missing permission, dependency down, partial completion?
- Data and migration. If any stored shape changes, is the migration named, and is the rollback named? If nothing changes, is that stated?
- Non-functional constraints. Auth and permission model, PII handling, performance budget, observability. Absent means unstated, not "not applicable"; "not applicable" must be written down.
- Test anchors. Does every acceptance criterion name at least one test that will prove it? For tickets, does every ticket name a test?
- Out of scope. Is there an explicit list of what this work will not do?
What to do with a gap
Each absent or partial becomes exactly one of:
- A question to the human, if the answer changes the design. Put it in the summary block under NEXT, one question per turn.
- An explicit assumption written into the artifact under
## Assumptions, if a sensible default exists. Assumptions are reviewed at the next gate.
Never silently fill a gap. Never leave a gap unrecorded.
Output
A table of the six questions with status and reference, the list of
questions and assumptions, then a record block with agent set to the agent
running this skill, claim GAPS: <n> or GAPS: 0, and one doc evidence
entry per gap.
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.