From aidd-refine
Verifies factual claims in text against authoritative sources using a cheapest-first cascade (project memory, codebase, web) and rewrites with footnote citations, hedging unverified claims.
How this skill is triggered — by the user, by Claude, or both
Slash command
/aidd-refine:05-fact-checkThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Verifies the factual claims inside a target text and rewrites it grounded in evidence. The skill extracts each verifiable claim, runs a cheapest-first verification cascade, and emits a rewritten answer where every confirmed claim carries a footnote citation and every unconfirmed claim is explicitly hedged. When sources disagree, both are reported rather than one being silently chosen.
Verifies the factual claims inside a target text and rewrites it grounded in evidence. The skill extracts each verifiable claim, runs a cheapest-first verification cascade, and emits a rewritten answer where every confirmed claim carries a footnote citation and every unconfirmed claim is explicitly hedged. When sources disagree, both are reported rather than one being silently chosen.
| # | Action | Role | Input |
|---|---|---|---|
| 01 | identify-claims | Extract verifiable factual claims from the text, classify each, drop opinion | target text |
| 02 | verify | Run the verification cascade per claim, produce a verdict and sources | claim list from 01 |
| 03 | report | Rewrite the text with footnote citations, hedge unverified, surface conflicts | verdict list from 02 |
Sequential skill: 01 → 02 → 03. No skipping. The router materializes the three actions as a task list on entry and closes each task only when its ## Test passes.
@references/verification-cascade.md.@references/claim-categories.md. Opinion, preference, and trivially-known statements are not claims and are skipped.(unverified - no source found).## Sources, and ## Unverified claims, nothing else. Internal mechanics never appear in the output: no cascade or tier trace (Cascade:, tier 1/2/3, miss, resolved), no category labels, no raw verdict words. State conclusions, not the process. Action 03 holds the exhaustive forbidden list.@references/claim-categories.md - locked taxonomy of verifiable claim categories with definition and example.@references/verification-cascade.md - the three-tier cascade, short-circuit rule, web-cost guardrail.@assets/report-template.md - rewritten-answer skeleton with the ## Sources footnote block.npx claudepluginhub ai-driven-dev/framework --plugin aidd-refineExtracts and verifies factual claims from PR copy or journalistic drafts, providing citations and warning on low certainty. Use before sending pitches or press releases.
Corroborates claims against external URLs by fetching content with smart-fetch or WebFetch, analyzing credibility, and recording findings as evidence or risks. Useful for verifying statements in code docs or research.
Produces grounded, citation-backed responses from source documents via direct quotes, uncertainty permission, and claim verification. Use for analyzing codebases, specs, or long documents.