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.
npx agentmods add instructions/turenlabs/batou/claude-mdgit clone --depth 1 https://github.com/turenlabs/batouWrote 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.
[](https://agentmods.dev/instructions/turenlabs/batou/claude-md)<a href="https://agentmods.dev/instructions/turenlabs/batou/claude-md"><img src="https://agentmods.dev/badge/instructions/turenlabs/batou/claude-md.svg" alt="Measured on agentmods" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.06718 | $0.06718 |
| Opus 5 | $0.03359 | $0.03359 |
| Sonnet 5 | $0.01344 | $0.01344 |
| Haiku 4.5 | $0.00672 | $0.00672 |
Grade A, and why
batou CLAUDE.md 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 4d 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.
How it starts
The opening of the file, as written. The whole thing — 275 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Batou - Code guard for your AI agents
Project Overview
Batou is a security scanner that runs as a Claude Code hook, analyzing code for vulnerabilities at write time. It intercepts Write, Edit, and NotebookEdit tool calls via PreToolUse (can block) and PostToolUse (provides hints) hooks.
Architecture
batou-core/cmd/batou/main.go Entry point — default: reads a Claude Code hook event from stdin, runs the scanner, outputs hints. Also: `batou scan <dir>` (repo scanner) and `batou findings` (findings-cache report)
batou-core/scanner/ Core scan orchestrator (concurrent rule execution + preprocessing)
batou-core/scanner/dirscan/ `batou scan DIR` subcommand: parallel directory walk over a shared in-memory call graph, JSONL or `--sarif` output, `--fail-on` CI gate, cross-file finalize pass
batou-rules/rules/ 45 rule categories (751 regex-based rules)
batou-core/ast/ Tree-sitter AST parsing (parser, query, filter, context)
batou-core/analyzer/ Language detection + 16 AST security analyzer packages
batou-core/taint/ Taint types + regex fallback engine (source -> sink tracking with sanitizers)
batou-core/taint/astflow/ Go-specific AST taint walker (uses go/ast, tracks channels/goroutines)
batou-core/taint/ssaflow/ Go SSA taint engine (golang.org/x/tools/go/ssa): def-use chains, cross-function fixpoint summaries, module-local cross-package builds. ON by default; opt out with BATOU_SSAFLOW=0
batou-core/taint/tsflow/ Tree-sitter taint walker (17 language configs; Zig's config is inert — no grammar)
batou-core/taint/languages/ Language-specific taint catalogs (17 languages, 73 files)
batou-core/hints/ Hint generation for Claude feedback (language-specific fix examples)
batou-core/graph/ Persistent call graph + interprocedural analysis
batou-core/hook/ Hook I/O (JSON stdin/stdout, exit codes)
batou-core/reporter/ Result formatting (block messages with CWE/OWASP refs)
batou-core/suppress/ Inline suppression parsing and matching (batou:ignore directives)
batou-core/findings/ Findings persistence + lifecycle tracking (new/recurring/fixed/suppressed)
batou-core/ledger/ Session audit logging
batou-core/testutil/ Test framework helpers
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.
- 4d ago First seen · 275 lines · 6,718 tokens per session scan A 46af54bc4fdd
batou CLAUDE.md is an instructions file published in the GitHub repository turenlabs/batou (5 stars, last pushed 1mo ago), licensed MIT. It adds 6,718 tokens to every session, about $0.0336 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 2026-08-31.
Other instructions, from other repositories
code-pathfinder CLAUDE.md
Instructions for shivasurya/code-pathfinder, covering claude.md, essential build commands, building the binary, running tests and linting.
stride-gpt AGENTS.md
AGENTS.md instructions for mrwadams/stride-gpt, covering agents.md, what this project is, repository layout, how the agentic analysis works and progressive disclosure pattern.
arcjet-js AGENTS.md
Instructions for arcjet/arcjet-js, covering agent guidance, examples live in arcjet/examples and integration work: review before a pr.
bex-security AGENTS.md
AGENTS.md instructions for bex-co/bex-security, covering keep it simple, avoid speculative defenses, public cli changes and public repository and pull requests.
ai-appsec AGENTS.md
Instructions for subodhkc/ai-appsec, covering agents.md — haiec agent security mcp (phase -1 → phase 18), 1. workspace layout, 2. architectural principles (apply to every phase), 3. hard constraints (never violate) and 4. phase discipline.
assay CLAUDE.md
Claude Code instructions for Rul1an/assay, covering assay - ai agent context, what is assay?, workspace structure, key commands and cli entry points.