agentleFS
Sign inStart free
wedabrowedabro/bro-skills/.agents/skills/speckit.clarify/SKILL.md

speckit.clarify

Skill3 starsChanged 45 days agoMITView on GitHub

Ambiguity Resolver - Detect, prioritize, and close specification ambiguity without inventing requirements.

We split this skill into the rules it gives an agent

5 rules. 0 are common, 3 are distinctive, and 2 are very specific to this project.

  • 0

    Common

    Most of the 3 edited versions we read keep these.

  • 3

    Distinctive

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Read spec.md, constitution, existing product conventions, and linked requirements.

    line 10
  2. Very specific:

    Update the spec with approved answers, assumptions, decision provenance, and [CLARIFIED] markers.

    line 11
  3. Distinctive:

    Ask at most three Critical questions in one batch.

    line 16
  4. Distinctive:

    No Critical ambiguity remains, every decision has source/provenance, and unresolved Important items are visible with owner and next action.

    line 21
  5. Distinctive:

    Do not change user intent, answer on the user's behalf, exceed three Critical questions, or edit implementation code.

    line 24

“Common” is measured: kept by at least 6 in 10 of the 3 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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Mission

Find uncertainties that could change architecture, cost, permissions, behavior, or acceptance, then resolve them with the smallest set of user questions.

Inputs and Outputs

  • Read spec.md, constitution, existing product conventions, and linked requirements.
  • Update the spec with approved answers, assumptions, decision provenance, and [CLARIFIED] markers; keep an ambiguity log when questions remain.

Protocol

  1. Scan for vague terms, missing boundaries, actors, permissions, error behavior, data ownership, non-functional targets, and contradictory IDs.
  2. Classify Critical (changes architecture/security/contract), Important (changes behavior/effort), or Minor (editorial only), citing the exact section.
  3. Ask at most three Critical questions in one batch; provide neutral options, trade-offs, and a recommended default without presuming approval.
  4. Auto-fix only behavior-neutral editorial issues. Record unanswered items as explicit assumptions or blockers; never silently invent limits or permissions.
  5. Apply approved answers to all affected scenarios, requirements, plan links, and acceptance criteria; re-scan for new conflicts.

Completion Gate

No Critical ambiguity remains, every decision has source/provenance, and unresolved Important items are visible with owner and next action.

Guard Rails

Do not change user intent, answer on the user's behalf, exceed three Critical questions, or edit implementation code.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

Reviews can't be read right now.

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.

Check your setup, free