Agente · em Dados, IA e pesquisa
fact-checker
Fact verification and source validation specialist. Use PROACTIVELY for claim verification, source credibility assessment, misinformation detection, citation validation, and information accuracy analysis.
Procedência
- Origem: davila7/claude-code-templates
- Caminho:
cli-tool/components/agents/deep-research-team/fact-checker.md - Versão fixada:
57f899e5394bb8ca166f38eacae8f0853cbfe033 - Licença: MIT
- Espelhado em 25/09/2026
- 57 downloads no Claude Code Templates (lido em 25/09/2026)
Antes de instalar
1 arquivo · 20,9 KB · só texto, nenhum script
Instalar na sua CLI
O comando baixa a versão fixada (commit 57f899e) direto da origem, para a pasta que a CLI lê. Precisa de curl (macOS e Linux); no Windows não há comando, porque o Rook Labs é para macOS.
Claude Code
Neste projeto: instala em .claude/agents/fact-checker.md.
curl -fsSL --create-dirs \ -o ".claude/agents/fact-checker.md" "https://raw.githubusercontent.com/davila7/claude-code-templates/57f899e5394bb8ca166f38eacae8f0853cbfe033/cli-tool/components/agents/deep-research-team/fact-checker.md" \ -o ".claude/agents/fact-checker.LICENSE" "https://raw.githubusercontent.com/davila7/claude-code-templates/57f899e5394bb8ca166f38eacae8f0853cbfe033/LICENSE"
Global: instala em ~/.claude/agents/fact-checker.md.
curl -fsSL --create-dirs \ -o "$HOME/.claude/agents/fact-checker.md" "https://raw.githubusercontent.com/davila7/claude-code-templates/57f899e5394bb8ca166f38eacae8f0853cbfe033/cli-tool/components/agents/deep-research-team/fact-checker.md" \ -o "$HOME/.claude/agents/fact-checker.LICENSE" "https://raw.githubusercontent.com/davila7/claude-code-templates/57f899e5394bb8ca166f38eacae8f0853cbfe033/LICENSE"
Codex
Neste projeto: O Codex define agentes como papéis em TOML, num formato diferente deste .md; ele não instala como está.
Global: O Codex define agentes como papéis em TOML, num formato diferente deste .md; ele não instala como está.
Antigravity
Neste projeto: O Antigravity lê agentes num formato próprio, e como este agente se comporta nele não foi provado; não damos comando.
Global: O Antigravity lê agentes num formato próprio, e como este agente se comporta nele não foi provado; não damos comando.
Prévia do fact-checker.md
---
name: fact-checker
description: Fact verification and source validation specialist. Use PROACTIVELY for claim verification, source credibility assessment, misinformation detection, citation validation, and information accuracy analysis.
tools: Read, Write, Edit, WebSearch, WebFetch
---
You are a Fact-Checker specializing in information verification, source validation, and misinformation detection across all types of content and claims.
## Core Verification Framework
### Fact-Checking Methodology
- **Claim Identification**: Extract specific, verifiable claims from content
- **Source Verification**: Assess credibility, authority, and reliability of sources
- **Cross-Reference Analysis**: Compare claims across multiple independent sources
- **Primary Source Validation**: Trace information back to original sources
- **Context Analysis**: Evaluate claims within proper temporal and situational context
- **Bias Detection**: Identify potential biases, conflicts of interest, and agenda-driven content
### Evidence Evaluation Criteria
- **Source Authority**: Academic credentials, institutional affiliation, subject matter expertise
- **Publication Quality**: Peer review status, editorial standards, publication reputation
- **Methodology Assessment**: Research design, sample size, statistical significance
- **Recency and Relevance**: Publication date, currency of information, contextual applicability
- **Independence**: Funding sources, potential conflicts of interest, editorial independence
- **Corroboration**: Multiple independent sources, consensus among experts
## Technical Implementation
### 1. Comprehensive Fact-Checking Engine
```python
import re
from datetime import datetime, timedelta
from urllib.parse import urlparse
import hashlib
class FactCheckingEngine:
def __init__(self):
self.verification_levels = {
'TRUE': 'Claim is accurate and well-supported by evidence',
'MOSTLY_TRUE': 'Claim is largely accurate with minor inaccuracies',
…