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/verginglabs/mcp/agents-mdgit clone --depth 1 https://github.com/verginglabs/mcpWhat 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 | $0.00836 | $0.00836 |
| Opus 5 | $0.00418 | $0.00418 |
| Sonnet 5 | $0.00167 | $0.00167 |
| Haiku 4.5 | $0.00084 | $0.00084 |
Grade A, and why
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 yesterday.
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 — 48 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Verging Labs — agent reference (ships in this package so it is never training-stale)
Verging Labs publishes versioned quality benchmarks of the tools AI agents depend on, measured on simulated real work rather than self-reported claims. Rankings are free; structured machine data is pay per call over x402 (payment is the authentication: no accounts, no API keys).
Live surfaces, always current:
https://verginglabs.com/llms.txt— site guide.https://verginglabs.com/llms-full.txt— the whole site as markdown in one fetch.- Every page has a markdown twin: append
.mdto its path, sendAccept: text/markdown, or add?mode=agent. - Streamable HTTP MCP endpoint with the same tools as this package plus site search:
https://verginglabs.com/mcp(manifest:https://verginglabs.com/.well-known/mcp.json). - OpenAPI contract:
https://verginglabs.com/openapi.json. Auth model:https://verginglabs.com/auth.md.
Tools in this package
fetch_memory_index (free)
Agentic Memory Index rankings: agent memory tools scored on real multi-session work.
- Input:
{ "provider"?: string }— optional name filter; omit for all rows. - Output:
structuredContentwithlive,index,version,released,basis,providersTotal, andproviders[](per row: index score with confidence intervals, per-kind pass rates, failure attribution, cost per 1k successful answers, time to ready, retention where published). - If no release is live the payload is an explicit
{ "error": "NOT_YET_LIVE" }, never invented rankings.
fetch_search_index (free)
Agentic Search Index rankings: web search and retrieval tools for AI agents. Same input and output contract as fetch_memory_index.
check_x402_endpoint (paid, x402)
Full graded verdict for an x402 endpoint before an agent pays it.
- Input:
{ "resource": string }— absolute https URL. - Output when paid:
decision(allow|warn|block|unknown),score,liveness,subscores,reasons, price/payTo/chain drift. - Output when unpaid:
{ "paymentRequired": true, "rawChallenge": <the x402 payment requirements> }. Payaccepts[0](USDC on Base) with any x402 client and retry the underlying REST call with theX-PAYMENTheader, or configure your x402 client for this server's calls. - Price: $0.035 per call as of 2026-08-04; the 402 challenge itself is always the live price.
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.
- yesterday First seen · 48 lines · 836 tokens per session scan A 4fa9766430ca
mcp AGENTS.md is an instructions file published in the GitHub repository verginglabs/mcp (0 stars, last pushed 29d ago), licensed MIT. It adds 836 tokens to every session, about $0.0042 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
spec-kit AGENTS.md
AGENTS.md instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.
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.
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).
langchain AGENTS.md
AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).
next.js AGENTS.md
Instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.