fact-check

Verify document claims against codebase and git history.

51|5|Updated Jan 25, 2026
One-click install
npx skills add https://github.com/LeeJuOh/claude-code-zero --skill fact-check-leejuoh
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: fact-check
Source: https://github.com/LeeJuOh/claude-code-zero/tree/main/plugins/vision-powers/skills/fact-check
Command: npx skills add https://github.com/LeeJuOh/claude-code-zero --skill fact-check-leejuoh

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill ensures that claims made in documents (like reports or plans) accurately reflect the actual codebase and its git history, correcting inaccuracies and providing a clear verification summary.

Core Features & Use Cases

  • Factual Verification: Compares statements in a document against source code and git logs.
  • In-Place Correction: Automatically fixes factual errors in numbers, names, and descriptions directly within the document.
  • Automated Reporting: Generates a summary of verification results, including confirmed, corrected, and unverifiable claims.
  • Use Case: After a project recap report is generated, use this Skill to verify that all mentioned commit hashes, file counts, and feature descriptions accurately match the repository's state.

Quick Start

Verify the factual accuracy of the report located at '~/claude-code-zero/reports/project-recap.html'.

Frequently Asked Questions about fact-check

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I verify factual claims in documentation against the codebase and git history?

Verifying factual claims in documentation against the codebase and git history requires cross-referencing document statements with source code and git logs. This process applies to auditing reports, plans, and documentation for accuracy in quantitative data, naming conventions, and behavioral descriptions.

Can I automatically fix factual errors in numbers and names directly within a report?

Automatically fixing factual errors in numbers and names directly within a report is supported through in-place correction. This functionality edits the document directly to correct inaccuracies in quantitative data and naming conventions based on codebase verification.

What is the best way to audit project recap reports for commit hash and file count accuracy?

Auditing project recap reports for commit hash and file count accuracy is best handled by verifying mentioned commit hashes, file counts, and feature descriptions against the repository's actual state. This ensures that quantitative data and temporal information match the git history.

Does documentation fact-checking work with standard file reading and bash commands for git operations?

Documentation fact-checking works with standard file reading and bash commands for git operations by requiring tools for reading files, globbing, grepping, editing, and executing bash commands. These tools enable file analysis and git history verification within the codebase.

How do I generate a summary of confirmed, corrected, and unverifiable claims after a report audit?

Generating a summary of confirmed, corrected, and unverifiable claims after a report audit is achieved through automated reporting. This functionality produces a clear verification summary detailing which factual claims match the codebase and which were corrected.