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/agent-hellboy/mcp-server-fuzzer/agents-mdgit clone --depth 1 https://github.com/Agent-Hellboy/mcp-server-fuzzerWrote 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/agent-hellboy/mcp-server-fuzzer/agents-md)<a href="https://agentmods.dev/instructions/agent-hellboy/mcp-server-fuzzer/agents-md"><img src="https://agentmods.dev/badge/instructions/agent-hellboy/mcp-server-fuzzer/agents-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 | $0.01403 | $0.01403 |
| Opus 5 | $0.00701 | $0.00701 |
| Sonnet 5 | $0.00281 | $0.00281 |
| Haiku 4.5 | $0.00140 | $0.00140 |
Grade A, and why
mcp-server-fuzzer AGENTS.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 — 159 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AI Agent Instructions
Branch Creation
Always create a feature branch before making any code changes:
git checkout main
git pull origin main
git checkout -b <descriptive-branch-name>
Name the branch after the feature or fix, for example:
oauth-client-credentials-authgit-branch-isolationfix-batch-auth-bypass
Never commit directly to main.
Commit Message Style
Follow the project's short imperative style — no trailing period:
Add <feature>
Fix <bug>
Refactor <component>
Update <thing>
Remove <thing>
Examples from this repo:
Add OAuth client credentials authAdd git branch isolation for AI agentsFix batch JSON-RPC auth bypass in test serverRefine reporting boundaries and result contracts
Lint and Tests
Run before every commit:
tox -e ruff # lint
tox -e tests -- <test paths> # unit tests
tox venv setup
tox creates a managed venv at .tox/tests/. Use its Python for anything that
needs project dependencies (running the server, the CLI, ad-hoc scripts):
# First run: tox creates the venv automatically
tox -e tests -- tests/unit/
# Use the tox Python directly for scripts / servers / CLI
.tox/tests/bin/python examples/test_server.py
.tox/tests/bin/python -m mcp_fuzzer --help
# Install extra deps into the tox venv (e.g. server deps)
.tox/tests/bin/pip install uvicorn mcp anyio starlette
Never use the system python3 or pip3 for project code — the repo uses a
managed environment and the system Python may lack required packages or be
protected by PEP 668.
Codebase Exploration — Use Graphify First
A knowledge graph of this repo lives in graphify-out/. Before exploring the code manually, query the graph:
# Ask anything about structure, coverage, relationships
graphify query "how does X work"
graphify query "what calls Y"
graphify path "AuthModule" "Database"
graphify explain "OAuthClientCredentialsAuth"
Graphify is faster than grepping and surfaces cross-file relationships the raw code doesn't make obvious. Only fall back to grep/Read when the graph answer is insufficient.
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 · 159 lines · 1,403 tokens per session scan A ae7a75826a98
mcp-server-fuzzer AGENTS.md is an instructions file published in the GitHub repository Agent-Hellboy/mcp-server-fuzzer (46 stars, last pushed 26d ago), licensed MIT. It adds 1,403 tokens to every session, about $0.0070 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.
Other instructions, from other repositories
director CLAUDE.md
Claude Code instructions for fdmtl/director, covering claude.md, product overview and development workflow, quality standards and workflow, engineering excellence and library documentation lookup.
Genesis-World-MCP AGENTS.md
AGENTS.md instructions for Gnidreve/Genesis-World-MCP, covering agents.md — cas genesisworld mcp, what this is, standing orders — read first, design pillars and source layout & conventions.
pixabay-mcp-server CLAUDE.md
Claude Code instructions for hanoak/pixabay-mcp-server, covering claude.md, project overview, tech stack (decided — do not relitigate without discussion), architecture & folder conventions and coding standards.
atelier-mcp copilot-instructions.md
Copilot instructions for anshrajore/atelier-mcp, covering run npm run sync-adapters to update, universal ui/ux rules, universal backend architecture rules and framework presets (available in skills/atelier/presets/).
mcp-foundry copilot-instructions.md
Copilot instructions for Mahima-Sanketh-Git/mcp-foundry, covering 🧭 instructions for copilot, 📋 readme structure (always follow this exact order), 📐 section-by-section rules, 1. project banner / hero and 2. table of contents.
mcp-server-clerk CLAUDE.md
Claude Code instructions for BalajiSriraman/mcp-server-clerk, covering claude.md — clerk mcp server, project overview, tech stack, project structure and key architectural concepts.