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/supermhel/fengarde/claude-mdgit clone --depth 1 https://github.com/supermhel/fengardeWrote 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/supermhel/fengarde/claude-md)<a href="https://agentmods.dev/instructions/supermhel/fengarde/claude-md"><img src="https://agentmods.dev/badge/instructions/supermhel/fengarde/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 | $0.01184 | $0.01184 |
| Opus 5 | $0.00592 | $0.00592 |
| Sonnet 5 | $0.00237 | $0.00237 |
| Haiku 4.5 | $0.00118 | $0.00118 |
Grade B, and why
fengarde CLAUDE.md scanned grade B with 1 finding 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.
Asks for rootmediumPrivilege escalation
A mod that escalates privileges can change anything on the machine, not only the project.
Ports: 6379 Redis (bus), 9200 OpenSearch, 5601 OpenSearch Dashboards, 8000 inventory API, 8080 FENGARDE alert console. On Linux/WSL2, OpenSearch needs `sudo sysctl -w vm.max_map_count=262144` or it crashes the JVM. How it starts
The opening of the file, as written. The whole thing — 74 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
FENGARDE — an open-source SIEM pipeline (Apache-2.0). Remote: https://github.com/supermhel/fengarde.git. Raw security logs are collected, normalized to OCSF, run through correlation rules, and indexed as alerts.
Read SSOT.md first for current status. It is the canonical status/roadmap index; other docs (especially docs/superpowers/specs/ and per-service INTERFACE.md files) may be historical, aspirational, or partially stale — SSOT.md says which is which. If a doc disagrees with SSOT.md, SSOT.md wins.
Commands
Everything below the Docker section is zero-infra (memory bus, in-memory store, stub LLM) — no Docker/Redis/OpenSearch needed. This is the normal dev loop.
make test # full CI gate: contract validator + every workstream's tests (= ./run_all_tests.sh)
make e2e # acceptance test: SSH brute-force burst -> real alert, idempotent under replay (tools/demo_e2e.py)
python tools/integration_e2e.py # pipeline smoke test: WS-1 -> WS-2 -> WS-4 -> WS-3 on one memory bus
# Single workstream test (each service has its own test_contract.py):
cd services/ws2-normalization && python test_contract.py
# Targeted suites (see run_all_tests.sh for the full list):
python services/ws4-detection/test_window.py
python services/shared/test_runner.py
Makefile uses python3; on Windows override with PYTHON=python or invoke scripts directly.
Docker stack (full live pipeline)
Docker Desktop must be started manually first — check with docker version once; if the engine is down, ask the user to start Docker Desktop, don't try to launch/poll it.
make preflight # doctor: checks vm.max_map_count, Docker RAM, free ports (required before first run)
make up # docker compose -f infra/docker-compose.yml up -d
make down # stop stack + remove volumes
Ports: 6379 Redis (bus), 9200 OpenSearch, 5601 OpenSearch Dashboards, 8000 inventory API, 8080 FENGARDE alert console. On Linux/WSL2, OpenSearch needs sudo sysctl -w vm.max_map_count=262144 or it crashes the JVM.
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 · 74 lines · 1,184 tokens per session scan B 3d39cb00c623
fengarde CLAUDE.md is an instructions file published in the GitHub repository supermhel/fengarde (11 stars, last pushed 6d ago), licensed Apache-2.0. It adds 1,184 tokens to every session, about $0.0059 per session on Opus 5. A static security scan graded it B with 1 finding (asks for root). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other instructions, from other repositories
selfhost-ai CLAUDE.md
Claude Code instructions for kossakovsky/selfhost-ai, covering claude.md, project overview, core architecture, key files and installation flow.
tutor-management-app CLAUDE.md
Instructions for t1112000/tutor-management-app, covering claude.md, commands, architecture, auth and page pattern.
homelab-monitor CLAUDE.md
Claude Code instructions for SikamikanikoBG/homelab-monitor, covering claude.md and conventions.
tutor-management-app AGENTS.md
Instructions for t1112000/tutor-management-app, covering repository guidelines, project structure & module organization, build, test, and development commands, coding style & naming conventions and testing guidelines.
piper CLAUDE.md
Claude Code instructions for piperbox/piper, covering claude.md, what this is, coding principles, 1. think before coding and 2. simplicity first.
homelab-mcp CLAUDE.md
Instructions for bjeans/homelab-mcp, covering claude development guide for homelab mcp, project overview, core philosophy, project structure and architecture patterns.