requirement-clarifier — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited requirement-clarifier (Agent Skill) and scored it 96/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 1 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 1 flagged
The text {match} tells the agent to skip the normal "ask the user first" gate. Used adversarially it removes the human-in-the-loop check before destructive or sensitive actions, turning a normally-gated agent into a fire-and-forget executor.
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
Prevent misbuilds without turning ordinary work into an interview.
Activate only when at least one unresolved choice could materially change:
For low-impact uncertainty, record one bounded assumption and continue. When quizme-mode is active, keep clarifying until the contract is complete.
Track these fields for each deliverable:
goalin_scopeout_of_scopeacceptance_criteriaconstraintsprohibitionsauthorityassumptionsmaterial_uncertaintiesClassify every requirement source as:
explicit_userhigher_priority_policyrepository_policynecessary_implicationbounded_assumptionoptional_enhancementNever present an assumption or enhancement as an explicit user requirement.
Ask only when uncertainty could cause meaningful rework, risk, or a different product outcome.
Always clarify before:
Before completion, map every deliverable and acceptance criterion to exactly one status:
satisfiedverifiedunverified with reasondeferred with authorizationblocked with blockerNo requirement may disappear during planning, implementation, or response composition.
For complex or multi-deliverable work, read task-contract-schema.md.
Keep the contract conversation-local by default. Write it to the repository only when:
--quizme --record is activeActivating this skill never grants file-write authority.
When clarification is required, provide:
When proceeding under assumptions, surface only assumptions that affect the result or residual risk.
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.