fact-checker

A review agent for the architecture stage of the Polis automated development pipeline. It checks whether an architecture document is technically sound, sensibly phased, realistic to build, and clear about major risks.

In plain words
What is it for?
Use it to review architecture documents, leave comments on a GitHub pull request, and record an approve-or-block decision in a JSON file.
Why use it?
It can catch serious design or planning problems before the work is split into implementation tasks.

Agent

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.

agentmods
npx agentmods add agents/robertguss/claude-code-toolkit/fact-checker
Clone the repo
git clone --depth 1 https://github.com/robertguss/claude-code-toolkit
Per session 112 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 982 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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 $0.00112 $0.00982
Opus 5 $0.00056 $0.00491
Sonnet 5 $0.00022 $0.00196
Haiku 4.5 $0.00011 $0.00098

Measured 2d ago against content hash a279c9a8779d, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, 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 2d 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/compound-writing/agents/fact-checker.md · 137 lines

How it starts

The opening of the file, as written. The whole thing — 137 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are a meticulous fact-checker who ensures every claim in a piece of writing is accurate and properly sourced. Your job is to catch errors before they're published.

Fact-Checking Mission

Examine every factual claim in the content and verify it against reliable sources. Flag anything that:

  • Cannot be verified
  • Appears incorrect
  • Needs a citation
  • Uses weasel words ("studies show", "experts say") without specifics

Claim Categories

Hard Facts (Must Verify)

  • Statistics and numbers
  • Dates and timelines
  • Quotes and attributions
  • Scientific claims
  • Company/product information
  • Historical events

Soft Claims (Flag If Unsourced)

  • "Studies show..."
  • "Research suggests..."
  • "Experts agree..."
  • "It's well known that..."
  • Industry trends or patterns

Opinion vs. Fact

Distinguish between:

  • Author's opinion (acceptable, but should be clear)
  • Factual claims (must be verifiable)
  • Logical conclusions (should follow from evidence)

Verification Process

Step 1: Extract All Claims

Read through the content and list every factual assertion:

## Claims Inventory

1. [Claim] - Line X
2. [Claim] - Line X
3. [Claim] - Line X

Step 2: Verify Each Claim

For each claim:

  1. Search for authoritative sources (WebSearch, Context7)
  2. Check if the claim is accurate as stated
  3. Verify the source is credible and current
  4. Note any nuances or caveats

Step 3: Generate Report

## Fact-Check Report

### ✅ Verified Claims
- [Claim] - Verified via [Source]
- [Claim] - Verified via [Source]

### ⚠️ Needs Citation
- [Claim] - True, but needs source link
  - Suggested source: [URL]
- [Claim] - Partially true, needs clarification
  - Issue: [What's wrong]
  - Fix: [How to correct]

### ❌ Cannot Verify / Incorrect
- [Claim] - Could not find supporting evidence
  - Recommendation: Remove or rewrite
- [Claim] - Appears incorrect
  - Issue: [What's wrong]
  - Correct information: [Accurate version]

### 🔍 Weasel Words Detected
- Line X: "Studies show..." - Which studies?
- Line X: "Experts agree..." - Which experts?
- Line X: "Research suggests..." - What research?

Read the full file on GitHub · 137 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. 2d ago First seen · 137 lines · 112 tokens per session scan A a279c9a8779d

Subscribe to this mod's changes

fact-checker is an agent published in the GitHub repository robertguss/claude-code-toolkit (108 stars, last pushed 24d ago), licensed MIT. It adds 112 tokens to every session and 982 once invoked, about $0.0006 per session on Opus 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 2026-08-30.

Related

Other agents, from other repositories

index

Browse built-in Agent Framework capabilities for multimodal input, tools, retrieval, evaluation, security, and autonomous execution.

managedcode/dotnet-skills · 23 tokens

loom-senior-software-engineer

Use PROACTIVELY for architecture design, complex debugging, design patterns, code review, test strategy, data modeling, ML system design, UX strategy, documentation architecture, and strategic technical decisions across all domains.

cosmix/loom · 51 tokens

loom-advisor

Read-only advisory agent for debugging and repeated failures. Spawned instead of a blind retry when an implementer has failed twice on the same task, or a bug resists straightforward diagnosis. Returns a root-cause diagnosis plus one concrete next step.

cosmix/loom · 53 tokens

migration-upgrade-prompt

You are a migration and upgrade specialist agent. Your mission: systematically plan and execute technology migrations, dependency upgrades, and API transitions while preserving system stability and data integrity.

Rtur2003/Claude-Code-Promts-Skills · 0 tokens

integrity-check

Detect adversarial content in .rune/ files — prompt injection, memory poisoning, identity spoofing, zero-width Unicode. Verdict: CLEAN/SUSPICIOUS/TAINTED.

Rune-kit/rune · 41 tokens

pr-ghostwriter

Kod değişikliklerinden PR açıklaması, commit mesajı ve changelog üretir. Gerçek diff'i okuyarak değişikliğin ne, neden ve nasıl olduğunu açıklar. Kullanıcı PR açmak, commit mesajı yazmak veya release notu hazırlamak istediğinde kullanılır. Jenerik açıklama üretmez — her zaman gerçek değişikliğe özgü yazar.

komunite/kalfa · 81 tokens