Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
Version History
Fix: --llm is optional
What This Skill Does
TruthCheck is a CLI tool that verifies claims, rates URL trustworthiness, and traces misinformation to its origin. It returns a TruthScore from 0 to 100 with a breakdown across publisher credibility, corroboration, and existing fact-checks. An LLM can be added optionally to improve content analysis accuracy.
It combines publisher reputation scoring, web corroboration, and optional LLM analysis in one command, removing the need to manually cross-reference multiple fact-checking sites.
When to use it
- Checking a viral social media claim before sharing it
- Verifying sources cited in a document or report
- Detecting hallucinated URLs in AI-generated content
- Tracing a news story back to its original source
- Assessing publisher credibility before citing an article
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: Fact-check this claim: 'Drinking coffee daily reduces Alzheimer's risk'
- 1Run `truthcheck verify "Drinking coffee daily reduces Alzheimer's risk" --llm gemini`
- 2Wait for the TruthScore and breakdown across publisher, content, corroboration, and fact-check dimensions
- 3Parse the score label and supporting evidence summary
- 4Report the verdict and which sources corroborate or contradict the claim
TruthScore: 65/100 (POSSIBLY TRUE) — some studies support an association, but evidence is inconclusive and no major health body has confirmed causation
Requirements
Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.