What is agentguard CLAUDE.md?

Repository instructions for AgentGuard, a local Go MCP server that checks code and dependencies for security issues. It scans for secrets and checks packages against OSV.dev, a public vulnerability database.

Use it when developing, testing, building, running, or distributing AgentGuard and its security-scanning tools.

It documents the correct build, test, run, and release commands while noting which checks need network access.

How shuaicongxiaomai/agentguard configures Claude Code

Reuse

Borrowing it

Nothing to install: this file belongs to shuaicongxiaomai/agentguard. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file

curl -O https://raw.githubusercontent.com/shuaicongxiaomai/agentguard/main/CLAUDE.md

Clone the repo

git clone --depth 1 https://github.com/shuaicongxiaomai/agentguard

Made for: Claude Code.

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.

README.md

[![agentmods](https://agentmods.dev/badge/instructions/shuaicongxiaomai/agentguard/claude-md/github.svg)](https://agentmods.dev/instructions/shuaicongxiaomai/agentguard/claude-md)

Your own site

<a href="https://agentmods.dev/instructions/shuaicongxiaomai/agentguard/claude-md"><img src="https://agentmods.dev/badge/instructions/shuaicongxiaomai/agentguard/claude-md/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.

Your own site · 80×15

<a href="https://agentmods.dev/instructions/shuaicongxiaomai/agentguard/claude-md"><img src="https://agentmods.dev/badge/instructions/shuaicongxiaomai/agentguard/claude-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 871 This file is loaded in full into every session.

When invoked 871 The same file — it is already loaded in full.

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.

Model	Per session	Once invoked
Fable 5.1	$0.00871	$0.00871
Opus 5	$0.00436	$0.00436
Sonnet 5	$0.00174	$0.00174
Haiku 4.5	$0.00087	$0.00087

Measured 8d ago against content hash f74841866994, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade A, and why

agentguard 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 8d 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.

CLAUDE.md · 59 lines

How it starts

The opening of the file, as written. The whole thing — 59 lines — stays where its author put it; the contents beside it link to each section on GitHub.

CLAUDE.md

This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.

What this is

AgentGuard is a single-binary MCP server (Go) that exposes three security tools to AI agents: scan_secrets (local regex + entropy), check_cve (single package via OSV.dev), and scan_dependencies (manifest-wide via OSV.dev). Free and open source; everything runs locally except CVE lookups, which hit the free public OSV.dev API.

Commands

main.go is at the repo root, not under cmd/ — the README's go build ./cmd/agentguard is stale.

go build -o agentguard .          # build (append .exe on Windows)
go test ./...                     # all tests
go test ./internal/agentguard -run TestScanSecrets -v   # single test
go vet ./...
./agentguard demo                 # run all three tools on sample inputs, zero config
./agentguard                      # stdio transport (default — MCP client launches the binary)
./agentguard serve :8080          # Streamable HTTP transport; also honors $PORT

Tests that hit check_cve / scan_dependencies paths reach OSV.dev over the network; the parser/secret tests are offline. go test covers only the pure parsing + secret-scan logic — there are no tests for the network (QueryOSV) or MCP-wiring layers.

Architecture

Two layers, kept strictly separate:

main.go — MCP wiring only. Defines the tool I/O structs (the jsonschema: struct tags generate the schema the agent sees), registers the three tools via mcp.AddTool, and builds human-readable text summaries. Both transports (StdioTransport, NewStreamableHTTPHandler) share one newServer().
internal/agentguard/ — all logic, no MCP dependency, so it's unit-testable in isolation:
- secrets.go — ScanSecrets: high-precision named patterns (AWS, GitHub, Anthropic, etc.) run first, then a generic key = "..." assignment regex gated by Shannon entropy ≥ 3.5 to cut false positives. Dedupes so a value caught by a named rule isn't re-reported as generic. Raw secret values are never returned — only masked (mask()).
- osv.go — QueryOSV: POSTs to https://api.osv.dev/v1/query, decodes into a trimmed Vuln. No API key.
- deps.go — ParseManifest dispatches on type (go.mod / package.json / requirements.txt) into per-format parsers, then ScanDependencies fans out to OSV concurrently (semaphore of 8, capped at maxDeps = 200). Parsers intentionally skip non-pinned specs (npm ranges/latest/URLs, pip non-==) since OSV needs an exact version.

Read the full file on GitHub · 59 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.

8d ago First seen · 59 lines · 871 tokens per session scan A f74841866994

Subscribe to this mod's changes

agentguard CLAUDE.md is an instructions file published in the GitHub repository shuaicongxiaomai/agentguard (0 stars, last pushed 2mo ago), licensed MIT. It adds 871 tokens to every session, about $0.0044 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.

Security

next.js AGENTS.md

AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.

vercel/next.js · 7,296 tokens

codex AGENTS.md

AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.

openai/codex · 5,153 tokens

vscode buildNext.instructions.md

Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).

microsoft/vscode · 6,785 tokens

agentguard: Instructions file for Claude Code