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/nvlang/verso-mcp/agents-mdgit clone --depth 1 https://github.com/nvlang/verso-mcpWrote 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/nvlang/verso-mcp/agents-md)<a href="https://agentmods.dev/instructions/nvlang/verso-mcp/agents-md"><img src="https://agentmods.dev/badge/instructions/nvlang/verso-mcp/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 | $0.02795 | $0.02795 |
| Opus 5 | $0.01398 | $0.01398 |
| Sonnet 5 | $0.00559 | $0.00559 |
| Haiku 4.5 | $0.00280 | $0.00280 |
Grade A, and why
verso-mcp 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 3d 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 — 204 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
Guidance for AI agents (and humans) asked to maintain or modify this repository.
This file follows the AGENTS.md convention — a plain
Markdown file of instructions that agentic coding tools read before working in a
repo. It is the canonical maintenance guide for verso-mcp; if anything
here conflicts with a guess or a habit, this file wins.
The owner wants this repository to need near-zero ongoing maintenance. Most "maintenance" (dependency bumps) is automated and should require no human or agent action. Read the relevant section below before changing anything, keep changes small, and do not invent work.
What this project is
verso-mcp is a small, single-purpose MCP
server: it lets an AI agent search and read documentation built with
Verso, Lean's documentation tool. It
exposes exactly four read-only tools — list_sites, list_kinds, search,
fetch_page — over stdio.
It is deliberately small. Resist scope creep. New tools, new dependencies, new config surface, and "frameworks" are almost always the wrong call. A change that makes the server bigger needs a strong, specific justification.
Repository layout
| Path | Purpose |
|---|---|
src/verso_mcp/server.py |
The entire server. One file on purpose. |
src/verso_mcp/__init__.py, __main__.py |
Thin entry-point shims. |
tests/ |
pytest suite — offline (test_core.py, test_fetch.py) plus the live canary (test_xref_contract.py). |
pyproject.toml |
Packaging, dependencies, tool config (ruff, pytest, commitizen). |
server.json |
MCP registry manifest. |
evaluation.xml |
10-question eval suite (mcp-builder format). |
.github/workflows/ |
CI, CodeQL, release, Scorecard, Dependabot auto-merge, xref-drift canary. |
The version number lives in one place — __version__ in server.py — and
is surfaced everywhere else dynamically (hatchling) or by cz bump. Never edit a
version string by hand.
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.
- 3d ago First seen · 204 lines · 2,795 tokens per session scan A e7ba246283f2
verso-mcp AGENTS.md is an instructions file published in the GitHub repository nvlang/verso-mcp (1 stars, last pushed 29d ago), licensed Apache-2.0. It adds 2,795 tokens to every session, about $0.0140 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
VCVio AGENTS.md
Instructions for Verified-zkEVM/VCVio, covering vcvio — ai agent guide, fast start, attribution, headers, and docstrings, module scopes and what this project is.
formal CLAUDE.md
Instructions for yamafaktory/formal, covering formal, rust, the files that judge changes, hints and lean.
lean-interact CLAUDE.md
Instructions for nasqret/lean-interact, covering claude.md — operating instructions for this repository, 0. start of every session, 1. the formalization loop, turn length, and deferred bookkeeping and when the proof is the interesting part.
leanforge-mcp AGENTS.md
AGENTS.md instructions for sandraschi/leanforge-mcp, covering agents.md -- leanforge-mcp, stack, repo layout, critical rules and lean subprocess.
leanforge-mcp CLAUDE.md
Claude Code instructions for sandraschi/leanforge-mcp, covering claude.md -- leanforge-mcp, what this repo does, key concepts, when working on agent.py and when working on leanclient.py.
verified-polygon-intersection CLAUDE.md
Instructions for schildep/verified-polygon-intersection, covering guideline on placement of definitions/theorems, network access and bash commands.