Agent Skills
› mtarcure/claude-vibe-squad
› claim-verification
claim-verification
GitHub用于对已完成交付物的事实声明（如数据、引用、预测等）进行核查，将其分解为独立主张并逐一验证真伪。
Trigger Scenarios
需要验证文本中事实准确性
审查内容中的统计数据或引用来源
Install
npx skills add mtarcure/claude-vibe-squad --skill claim-verification -g -y
SKILL.md
Frontmatter
{
"name": "claim-verification",
"audience": "specialist",
"description": "Use to fact-check a finished deliverable's stated facts — its statistics, quotations, named sources, dates, and forecasts — by decomposing them into load-bearing factual claims and matching each to the exact source span that confirms or refutes it (Hard Rule 8 truth gate). Scope is whether what the text asserts as fact is accurate — the truth of the content itself, not whether a task's tooling actually ran. Route web facts to a grounding worker."
}
Claim Verification
Decompose content into load-bearing claims and verify each against evidence (Hard Rule 8 truth gate).
Steps
- Decompose the content into discrete load-bearing claims.
- Classify each claim:
fact | quote | calculation | forecast | opinion | inference. - Map each claim to the exact evidence span that would confirm it.
- Verify against sources; for web claims, route a grounding worker (Gemini grounding or research handoff) that returns a typed evidence bundle, then adjudicate it.
- Reproduce any arithmetic and check units/internal consistency.
- Return a per-claim verdict:
supported | unsupported | unverifiable.
Acceptance
- Every load-bearing claim is classified and mapped to evidence.
- "Unverifiable" is distinguished from "false"; a model cutoff is never treated as verification.
- No PASS while a load-bearing claim remains unverifiable.
Version History
- d5262e2 Current 2026-09-11 11:15
