Audits a spec.md or slice file for ambiguities across six defined categories and generates prioritized questions.
npx skills add https://github.com/ramboz/jig --skill clarify-ramboz
Please help me install this Agent Skill. Skill: clarify Source: https://github.com/ramboz/jig/tree/main/skills/clarify Command: npx skills add https://github.com/ramboz/jig --skill clarify-ramboz
Ambiguities in a Jig spec slow teams and introduce risk; this skill provides a lightweight, repeatable ambiguity audit that surfaces gaps before reviews.
Run clarify on a draft spec to surface ambiguities and append a Clarifications section.
High-intent search queries and answers about installing and using this skill.
To identify ambiguities in a draft spec before review, run a structured six-category audit that analyzes the document and generates prioritized questions to append to a Clarifications section.
A spec ambiguity audit should cover Scope & Boundaries, Acceptance Criteria Testability, Dependencies & Blockers, Non-functional Requirements, Edge Cases & Failure Modes, and Terminology Consistency.
Yes, you can run an ambiguity scanner on a single slice file independently. The standalone baseline processes one document at a time without deferring to other workflow steps.
A spec audit generates up to five prioritized questions written verbatim. These questions are appended to a Clarifications section alongside a coverage summary for audit traceability.
The best way to track spec readiness across cross-skill workflows is to use a coverage summary with Clear, Partial, or Missing ratings to guide Ready-for-Review decisions and output a Q&A trace.
Jig spec clarification does not require external workflow dependencies. The audit operates as a standalone baseline to surface gaps and append clarifications directly to the document.
rambozAudit a jig spec against related artifacts and emit a markdown findings report.
amoutiersResolve ambiguous spec items by generating clarifying questions and updating the document in place.
XinheLIUCompare spec.md, plan.md, and tasks.md to produce a read-only Markdown audit report.
nholder88Identify hidden assumptions, unknowns, and logical gaps in planning artifacts.
