spec-kit-clarify
Resolve ambiguities in a feature spec through targeted Q&A. Run after spec-kit-specify when open questions remain. Updates spec.md incrementally.
How do I install this agent skill?
npx skills add https://github.com/aircury/ai-framework --skill spec-kit-clarifyIs this agent skill safe to install?
- Gen Agent Trust Hubpass
The skill facilitates the resolution of functional and technical ambiguities in documentation through a structured questioning process. It performs directory scans, file reads, and updates to local markdown files, which are consistent with its stated purpose of managing project specifications.
- Socketpass
No alerts
- Snykpass
Risk: LOW · No issues
What does this agent skill do?
Resolve ambiguities in a feature spec through focused, sequential questioning.
Input: Optionally specify a change name. If omitted, infer from context or ask.
Steps
-
Select the change
Auto-select if only one directory exists under
specs/changes/. Otherwise ask. -
Load
specs/changes/<name>/spec.mdIf it does not exist, prompt the user to use the
spec-kit-specifyskill first. -
Scan for ambiguities
Examine the spec for issues across these categories (in priority order):
- Functional scope — unclear boundaries or behavior
- Data models — undefined entities, states, or transitions
- UX flows — missing user journeys or edge cases
- Non-functional — unspecified performance, security, or scale requirements
- Integrations — undefined external dependencies
- Terminology — ambiguous or inconsistent domain language
Generate up to 5 questions total. Skip categories with no ambiguity.
-
Ask one question at a time
Format each question as multiple-choice (2–5 options) or a short answer (≤5 words).
Only ask a question if the answer materially impacts architecture, data modelling, testing, UX, operations, or compliance.
After each answer, update
spec.mdbefore asking the next question. -
Finalise
After all questions are resolved (or the 5-question limit is reached), update the spec to:
- Remove resolved
[NEEDS CLARIFICATION]markers. - Flag any remaining unresolved items explicitly.
Output
After completing all questions:
- Confirm how many ambiguities were resolved.
- List any items deferred (still open).
- Next step: "Use the
spec-kit-planskill to create the technical implementation plan."
Guardrails
- Never ask more than 5 questions.
- Always save atomically after each answer — never batch updates.
- If all open questions are already resolved, report that and suggest the
spec-kit-planskill. - Never introduce implementation details into the spec during clarification.
How can the creator link this skill?
Add the canonical catalog link to the repository README so users can inspect current installs and available audits. The publishing guide covers the complete discovery path.
<a href="https://skillzs.dev/skills/aircury/ai-framework/spec-kit-clarify">View spec-kit-clarify on skillZs</a>