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/ericovis/cervo/claude-mdgit clone --depth 1 https://github.com/ericovis/cervoWrote 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/ericovis/cervo/claude-md)<a href="https://agentmods.dev/instructions/ericovis/cervo/claude-md"><img src="https://agentmods.dev/badge/instructions/ericovis/cervo/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 | $0.05831 | $0.05831 |
| Opus 5 | $0.02916 | $0.02916 |
| Sonnet 5 | $0.01166 | $0.01166 |
| Haiku 4.5 | $0.00583 | $0.00583 |
Grade B, and why
cervo 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 today.
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.
handed over with `:U`). Host-side, the runtimes still run as root. If a How it starts
The opening of the file, as written. The whole thing — 364 lines — stays where its author put it; the contents beside it link to each section on GitHub.
cervo
A demo app for managing static website hosting on a shared VPS, built as an MCP server with FastMCP. Claude Code is the AI/chat interface used to exercise and test the server's tools during development. The whole environment — dev and production alike — runs from docker-compose: caddy is the front door, a worker process runs deployments, and mailcatcher stands in for SMTP in development.
Running
Prerequisite: Docker (with Compose). uv is used for lint/format and can run the test suite directly.
bin/dev # docker compose up -d — the whole environment
The stack is four services, all sharing the .data volume at /mnt/data:
app— the MCP server and the public website, one service: uvicorn withWEB_CONCURRENCYworker processes (default 1) importingcervo.asgi— MCP runs stateless, so any worker serves any request. Publishes no port; caddy reverse-proxies it athttp://localhost— the homepage (with docs at/docs) — and the MCP endpoint stayshttp://localhost/mcp.worker— the job worker (cervo-worker): one container runningWORKER_CONCURRENCYpolling threads (default 1) — claiming is atomic in the database, so more threads never double-run a job. Boots first, creates the database tables, and renders the initial Caddyfile (once, before any thread polls).caddy— the front door on ports 80/443. Runs on the generated/mnt/data/Caddyfileand serves the sites athttp://{slug}.localhost. Its unauthenticated admin API (caddy:2019) is reachable only on the compose network. On a fresh checkout it restarts until the worker first renders the Caddyfile — that's therestart: unless-stoppeddoing its job, not a bug.mail— mailcatcher (SMTP on 1025, web UI at http://localhost:1080). Development mail — the sign-in codes — goes here, not to real SMTP.
Every cervo container runs unprivileged inside: the image bakes in a
cervo user (uid 1000) that owns /app and /mnt/data, and caddy runs
as the same uid with a sysctl re-allowing :80/:443 (its dev state goes to
/tmp — certificates only exist in production, where its volumes are
handed over with :U). Host-side, the runtimes still run as root. If a
.data directory written by an older root-run stack refuses writes,
delete it (dev data is disposable) or chown it to uid 1000.
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.
- today Changed · +32 lines · +590 tokens per session 4cff590944dd
- 3d ago First seen · 332 lines · 5,241 tokens per session scan B fb5d23514dab
cervo CLAUDE.md is an instructions file published in the GitHub repository ericovis/cervo (0 stars, last pushed 2d ago), licensed MIT. It adds 5,831 tokens to every session, about $0.0292 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 2026-08-31.
Other instructions, from other repositories
blockrun-mcp AGENTS.md
AGENTS.md instructions for BlockRunAI/blockrun-mcp, covering blockrun mcp, commands, project structure, key dependencies and install in codex.
openrouter-mcp-multimodal AGENTS.md
AGENTS.md instructions for stabgan/openrouter-mcp-multimodal, covering agent instructions, before you ship, releasing (read this before publishing), short version and version files (must all match package.json).
intervals-icu-mcp CLAUDE.md
Instructions for hhopke/intervals-icu-mcp, covering claude.md, project overview, development commands, architecture (quick reference) and tool categories.
ai-toolkit AGENTS.md
AGENTS.md instructions for pipefy/ai-toolkit, covering repository guidelines, documentation map, project structure, import namespace migration: pipefysdk → pipefy and src/pipefysdk/init.py (transitional shim).
flyto-core CLAUDE.md
Claude Code instructions for flytohub/flyto-core, covering claude notes, cross-agent handoff and shared code intelligence.
Plonk AGENTS.md
AGENTS.md instructions for ostapondo/Plonk, covering agent rules, layout, adding a module, build & verify and code style.