fact-check-code

fact-check-code is a command for Claude Code from bengous/claude-code-plugins. It costs 13 tokens per session (458 once invoked), scanned A, original, MIT.

A codebase-checking command that tests whether a stated claim matches the actual implementation.

In plain words
What is it for?
Use it to search relevant files, consult external documentation when needed, and report evidence with file references.
Why use it?
It helps catch assumptions based only on filenames, comments, or incomplete searches.

Command for Claude Code

Written for Claude Code: allowed-tools in frontmatter. Also seen: model in frontmatter.

Part of the research-tools plugin — 1 skill, 4 commands, 2 agents shipped together

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 commands/bengous/claude-code-plugins/fact-check-code
Clone the repo
git clone --depth 1 https://github.com/bengous/claude-code-plugins

Made for: Claude Code.

Or install research-tools, the plugin that ships this one along with the rest of its 1 skill, 4 commands, 2 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-check-code

README.md
[![agentmods](https://agentmods.dev/badge/commands/bengous/claude-code-plugins/fact-check-code.svg)](https://agentmods.dev/commands/bengous/claude-code-plugins/fact-check-code)
Your own site
<a href="https://agentmods.dev/commands/bengous/claude-code-plugins/fact-check-code"><img src="https://agentmods.dev/badge/commands/bengous/claude-code-plugins/fact-check-code.svg" alt="Measured on agentmods" height="20"></a>
Per session 13 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 458 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 5.1 $0.00013 $0.00458
Opus 5 $0.00006 $0.00229
Sonnet 5 $0.00003 $0.00092
Haiku 4.5 $0.00001 $0.00046

Measured 2d ago against content hash 8ab5616f361b, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

fact-check-code 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.

research-tools/commands/fact-check-code.md · 62 lines

How it starts

The opening of the file, as written. The whole thing — 62 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Fact Check (Code)

Verify the following claim about this codebase:

Process

  1. Search the codebase - Use Grep and Glob in parallel to find relevant files and patterns
  2. Read and verify - Examine the actual implementation, not just file names or comments
  3. Consult external sources - Use WebSearch/WebFetch if the claim involves library behavior or standards
  4. Report - Present findings with specific file references

<parallel_search_guidance> Run multiple searches in parallel: search for the claimed behavior, related function/class names, and configuration files simultaneously. Check multiple potential locations since code organization varies. </parallel_search_guidance>

Output Format

Verdict

<verdict_definitions>

  • Confirmed: Code evidence directly supports the claim
  • Partially True: Claim is accurate in some aspects but incomplete or contains inaccuracies
  • Incorrect: Code evidence contradicts the claim
  • Unable to Verify: Relevant code not found, or evidence is inconclusive </verdict_definitions>

Evidence

  • Cite specific files and line numbers: path/to/file.ts:42
  • Quote relevant code snippets when helpful
  • Explain discrepancies between claim and actual implementation

Sources

  • List files examined with line references
  • Include web sources if consulted for library/API behavior

Guidelines

  • Always search before concluding - never rely on assumptions or prior knowledge
  • Check multiple locations using different search terms (the code might use unexpected naming)
  • Distinguish "not found after thorough search" from "confirmed absent" - explain your search strategy
  • State your interpretation if the claim is ambiguous before searching

Read the full file on GitHub · 62 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 · 62 lines · 13 tokens per session scan A 8ab5616f361b

Subscribe to this mod's changes

fact-check-code is a command published in the GitHub repository bengous/claude-code-plugins (4 stars, last pushed 4d ago), licensed MIT. It adds 13 tokens to every session and 458 once invoked, about $0.0001 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 2026-09-03.

Related

Other commands, from other repositories