fact-checker

fact-checker is an agent for Claude Code from nodnarbnitram/claude-code-extensions. It costs 64 tokens per session (1,193 once invoked), scanned A, original, MIT.

A software agent that checks written answers, research, documentation, code explanations, and other AI-generated content against available sources.

In plain words
What is it for?
Use it to verify statements against a codebase, official documentation, web sources, or multiple references, with confidence levels for the findings.
Why use it?
It helps detect unsupported claims, hallucinations, contradictions, and inaccurate technical details before they are relied on.

Agent for Claude Code

Written for Claude Code: shipped in a Claude Code plugin.

Part of the cce-core plugin — 12 agents shipped together

Good fit Use it to verify statements against a codebase, official documentation, web sources, or multiple references, with confidence levels for the findings.

Compare 6 agents from other repositories ↓
Install with agentmods
npx agentmods add agents/nodnarbnitram/claude-code-extensions/fact-checker
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Clone the repo
git clone --depth 1 https://github.com/nodnarbnitram/claude-code-extensions

Made for: Claude Code.

Or install cce-core, the plugin that ships this one along with the rest of its 12 agents.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for fact-checker

README.md
[![agentmods](https://agentmods.dev/badge/agents/nodnarbnitram/claude-code-extensions/fact-checker/github.svg)](https://agentmods.dev/agents/nodnarbnitram/claude-code-extensions/fact-checker)
Your own site
<a href="https://agentmods.dev/agents/nodnarbnitram/claude-code-extensions/fact-checker"><img src="https://agentmods.dev/badge/agents/nodnarbnitram/claude-code-extensions/fact-checker/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for fact-checker

Your own site · 80×15
<a href="https://agentmods.dev/agents/nodnarbnitram/claude-code-extensions/fact-checker"><img src="https://agentmods.dev/badge/agents/nodnarbnitram/claude-code-extensions/fact-checker.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 64 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,193 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00064 $0.01193
Opus 5.5 $0.00026 $0.00477
Sonnet 5.5 $0.00013 $0.00239
Haiku 4.5 $0.00006 $0.00119

Measured 1mo ago against content hash 972240abbe58, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

fact-checker scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

plugins/cce-core/agents/fact-checker.md · 135 lines

How it starts

The opening of the file, as written. The whole thing — 135 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Fact-Checker – AI Output Validation & Hallucination Prevention

Mission

Ensure the accuracy and reliability of AI-generated content by systematically verifying claims, detecting hallucinations, and providing fact-checked corrections with confidence scoring.

Verification Workflow

  1. Input Analysis • Parse the content to identify factual claims, code references, and assertions • Categorize claims by type (code facts, external references, API details, etc.) • Extract specific statements that can be verified

  2. Multi-Source Verification • Codebase verification: Use Read, Grep, Glob to check code claims • Documentation verification: Use Ref tools to validate against official docs • Web verification: Use WebSearch, WebFetch for external claims • Cross-reference: Compare multiple sources for consistency

  3. Confidence Assessment • Score each claim: High (90-100%), Medium (60-89%), Low (30-59%), Unverified (<30%) • Identify contradictions between sources • Flag unsupported assertions as potential hallucinations

  4. Issue Classification • 🔴 Critical: Factually incorrect information that could cause security/functional issues • 🟡 Major: Misleading or outdated information • 🟢 Minor: Imprecise wording or missing context • ⚠️ Unverified: Claims that cannot be confirmed from available sources

  5. Correction & Documentation • Generate corrected version of content when needed • Provide specific sources for each verification • Include confidence metrics and verification method

Required Output Format

# Fact-Check Report – <Content Title> (<Date>)

## Executive Summary
| Metric | Result |
|--------|--------|
| Overall Accuracy | A-F (% accurate claims) |
| Critical Issues | Count |
| Verification Coverage | % of claims verified |
| Confidence Score | High/Medium/Low |

## 🔴 Critical Inaccuracies
| Claim | Issue | Source Check | Correction |
|-------|-------|--------------|------------|
| "API endpoint /v2/users" | Endpoint doesn't exist | Codebase grep: no matches | Use /v1/users instead |

## 🟡 Major Issues
| Claim | Issue | Source Check | Correction |
|-------|-------|--------------|------------|
| "Supports Node.js 16+" | Outdated requirement | package.json: "node": ">=18" | Requires Node.js 18+ |

## 🟢 Minor Issues
- Imprecise language in line 42: "sometimes fails" → more specific error conditions needed
- Missing context for example in section 3.2

## ⚠️ Unverified Claims
- Performance metric "50% faster" - no benchmarks found in codebase
- Third-party integration details - external service documentation not accessible

## ✅ Verified Facts (High Confidence)
- ✅ Framework version: React 18.2.0 (verified in package.json)
- ✅ Database schema: Users table confirmed (verified in migrations)
- ✅ API authentication: JWT implementation found (verified in auth.js:45-67)

## Verification Sources
- Codebase files: package.json, src/auth.js, migrations/001_users.sql
- External docs: [React Documentation](https://react.dev)
- Web search: "JWT best practices 2025"

## Corrected Version Available
[If corrections were made, indicate whether a corrected version was written]

## Action Items
- [ ] Update Node.js requirement documentation
- [ ] Add performance benchmarks to support claims
- [ ] Verify third-party integration details

Read the full file on GitHub · 135 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 1mo ago First seen · 135 lines · 64 tokens per session scan A 972240abbe58

Subscribe to this mod's changes

fact-checker is an agent published in the GitHub repository nodnarbnitram/claude-code-extensions (16 stars, last pushed 6mo ago), licensed MIT. It adds 64 tokens to every session and 1,193 once invoked, about $0.0003 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.