agentleFS
mtarcuremtarcure/claude-vibe-squad/.agents/skills/claim-verification/SKILL.md

claim-verification

Skill220 starsChanged 19 days agoMITView on GitHub

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.

7 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

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

  • 0

    Common

    Most of the 5 edited versions we read keep these.

  • 3

    Distinctive

    This author's own choices. Worth a look.

  • 0

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Verify against sources; for web claims, route a grounding worker (Gemini grounding or research handoff) that returns a typed evidence …

    line 15
  2. Distinctive:

    Every load-bearing claim is classified and mapped to evidence.

    line 20
  3. Distinctive:

    No PASS while a load-bearing claim remains unverifiable.

    line 22

“Common” is measured: kept by at least 6 in 10 of the 5 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

Claim Verification

Decompose content into load-bearing claims and verify each against evidence (Hard Rule 8 truth gate).

Steps

  1. Decompose the content into discrete load-bearing claims.
  2. Classify each claim: fact | quote | calculation | forecast | opinion | inference.
  3. Map each claim to the exact evidence span that would confirm it.
  4. Verify against sources; for web claims, route a grounding worker (Gemini grounding or research handoff) that returns a typed evidence bundle, then adjudicate it.
  5. Reproduce any arithmetic and check units/internal consistency.
  6. 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.

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.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

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.