Blind Fact Check — Data & AI agent for Claude Code
Zero-context blind review of a single chapter for factual errors and AI-template traces.
How to install Blind Fact Check
Installs to ~/.claude/agents/thefooliman-multi-agent-academic-paper-workflow-blind-fact-check.md
mkdir -p ~/.claude/agents && curl -fsSL https://raw.githubusercontent.com/thefooliman/multi-agent-academic-paper-workflow/HEAD/agents/blind-fact-check.md -o ~/.claude/agents/thefooliman-multi-agent-academic-paper-workflow-blind-fact-check.md Restart Claude Code, or start a new session, for it to be picked up.
What Blind Fact Check does
name: blind-fact-check description: Zero-context blind review of a single chapter for factual errors and AI-template traces. Receives only the chapter text. Use after self-review.
You are the Blind Fact-Check Agent. You have zero prior context and zero pre-read memory; you receive only the current chapter text and evaluate it in isolation.
Scope
- Factual errors: unsupported claims, implicit fake information, logical fallacies.
- Template homogeneity: high-density generic openings,
Alternatives in Data & AI
- Credential Hunter — Autonomous credential-attack pipeline runner 4.5k ★
- Info Disclosure — Information Disclosure specialist (H1 #18, CWE-200/209/215/538/668/798) 812 ★
- Migration Agent — Creates safe, reversible database migrations with proper indexes, constraints, and zero-downtime strategies 652 ★
Full documentation available on GitHub
View Source RepositoryRelated Agents
Moneydive Fact Checker
Verifies every factual claim in an equity intelligence report against source data. Cross-checks peers, financi
Prose Review
Reviews prose for AI patterns, banned phrases, voice drift, and structural issues. Dispatched by the copydesk
PDF Debugger
Debug PDF processing issues — OCR failures, text extraction problems, AI misclassification, and renaming error
Book Chapter Writer
AI 교육 책 챕터 집필 전문가. 기존 교육 자료를 독자 친화적이고 흥미로운 책 형식으로 재구성. '챕터 집필', '책 집필' 요청 시 사용.
AI Plan
Relentless interrogator and principal delivery architect. Entry point for non-trivial work: runs discovery, ex
Council Orchestrator
Coordinates the LLM Council — spawns five parallel advisors, collects blind responses, and synthesizes through
Related Skills
Dotnet Refactor
Phase-gated refactoring/design loop - session-blind subagents build the map and traces, empirical probes prece
Opik Openclaw
🦞 Official plugin for OpenClaw that exports agent traces to Opik. See and monitor agent behaviour, cost, toke
Root Cause Tracing
Use when errors occur deep in execution and you need to trace back to find the original trigger - systematical