Claim Source Auditor
Objective
Turn a draft or claim set into a traceable evidence map. Separate what is proven, what is contradicted, what is not found in the searched source set, and what still needs manual review.
Directives
| Parameter | Requirement |
|---|---|
| Formatting | Table-first; each claim gets a discrete status row |
| Tone | Factual, restrained, audit-style |
| Scope | Verification only; do not rewrite the whole narrative unless the verification results require targeted remediation |
| Grounding | Every verified or conflicting result must trace to a named file, source location, or quoted passage |
| Defaults | Use the fixed status taxonomy and keep unsourced claims unresolved rather than downgraded into speculation |
| Fallback | If search or extraction is partial, return the searched scope, likely matches, and manual-review flags instead of overstating certainty |
| Triggering | Use for source mapping, quote verification, claim checking, contradiction audits, and pre-release fact review |
| Restrictions | Do not treat absence as falsity without naming the searched scope; do not paraphrase a quote as verified unless the exact wording or a faithful equivalent is present |
Workflow
- Define the audit scope: claims list, searched sources, and whether exact quotes are required.
- Break compound paragraphs into individual claims.
- Search the provided source set for exact wording first, then faithful supporting context.
- Assign one status per claim:
verified,verified in broader bundle,conflicting,not found in searched sources, ormanual review needed. - Cite the exact source location when available.
- Separate remediation notes from evidence status so unsupported claims are easy to fix or remove.
Resources
| Resource | Create When | Notes |
|---|---|---|
scripts/ |
A claim-extraction or source-map helper exists | Optional |
references/ |
A source hierarchy or record legend is needed | Useful for large evidence sets |
assets/ |
A reusable audit table template is needed | Optional |
agents/openai.yaml |
The skill should appear cleanly in the UI | Keep the UI prompt focused on verification |
Preferred status meanings:
verified: supported in the exact searched packetverified in broader bundle: supported locally, but outside the narrow packet under reviewconflicting: the source materially contradicts the draft claimnot found in searched sources: not located in the named search scopemanual review needed: likely present, but extraction, image content, or ambiguity blocks automatic confirmation
Validation
- Verify that every claim row names the searched source scope.
- Verify that every
verifiedorconflictingrow includes a concrete citation or file reference. - Verify that every
not foundrow distinguishes source absence from search limitation. - Remove blended statuses and duplicated claims.
- Run
scripts/quick_validate.py <path/to/skill-folder>before delivery.