What is hf-mcp-server AGENTS.md?

A repository guide for a TypeScript monorepo containing an MCP server, its shared libraries, a web interface, and Python end-to-end tests. A monorepo stores multiple related packages in one repository; MCP is a standard way for AI assistants to use tools.

It helps with server and web-interface changes, shared MCP tools, package-specific tests, formatting, type checks, and full local runs.

It shows where each package lives and lists the commands for building, checking, testing, and running the whole project.

How huggingface/hf-mcp-server configures Codex and OpenCode

About the project

Hugging Face MCP Server is a Model Context Protocol server that connects language models and coding agents to the Hugging Face Hub and Gradio applications. Users install it in supported agent environments to access configured Hugging Face tools and Spaces. The catalogue instruction describes how to use this server.

huggingface/hf-mcp-server · 310 stars · on GitHub · huggingface.co

Reuse

Borrowing it

Nothing to install: this file belongs to huggingface/hf-mcp-server. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file

curl -O https://raw.githubusercontent.com/huggingface/hf-mcp-server/main/AGENTS.md

Clone the repo

git clone --depth 1 https://github.com/huggingface/hf-mcp-server

Made for: Codex, OpenCode.

Wrote 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.

README.md

[![agentmods](https://agentmods.dev/badge/instructions/huggingface/hf-mcp-server/agents-md/github.svg)](https://agentmods.dev/instructions/huggingface/hf-mcp-server/agents-md)

Your own site

<a href="https://agentmods.dev/instructions/huggingface/hf-mcp-server/agents-md"><img src="https://agentmods.dev/badge/instructions/huggingface/hf-mcp-server/agents-md/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/instructions/huggingface/hf-mcp-server/agents-md"><img src="https://agentmods.dev/badge/instructions/huggingface/hf-mcp-server/agents-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 604 This file is loaded in full into every session.

When invoked 604 The same file — it is already loaded in full.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

Origin original No closer match found in the catalogue.

Token cost

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 5.1	$0.00604	$0.00604
Opus 5.5	$0.00242	$0.00242
Sonnet 5.5	$0.00121	$0.00121
Haiku 4.5	$0.00060	$0.00060

Measured 1mo ago against content hash 06ee4a54fe08, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

hf-mcp-server 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 1mo 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.

Origin

Copies of this mod

1 near-identical copy found in the catalogue:

hf-mcp-server AGENTS.md — 100% identical, 9 lines differ

AGENTS.md · 47 lines

How it starts

The opening of the file, as written. The whole thing — 47 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Repository Guidelines

Subagents

Project Structure & Module Organization

packages/app: MCP server plus management web UI. Server code lives in packages/app/src/server, UI in packages/app/src/web, and tests in packages/app/test.
packages/mcp: shared MCP tools/library consumed by the server. Source is in packages/mcp/src, with tests in packages/mcp/test and a few *.test.ts files co-located in src.
packages/e2e-python: Python-based end-to-end harnesses and fixtures.
Root-level docs and support files live in docs/, docs-internal/, scripts/, spec/, and Dockerfile.

Build, Test, and Development Commands

pnpm install: install workspace dependencies (Corepack-managed pnpm).
pnpm build: build all workspace packages.
pnpm dev: watch the MCP package and run the server with HMR.
pnpm start: start the production server from packages/app.
pnpm lint, pnpm typecheck, pnpm format, pnpm test: run repo-wide checks.
Package-specific tests: pnpm -C packages/app test or pnpm -C packages/mcp test.
Full local pipeline: pnpm run buildrun (clean, build, lint, test, start).

Coding Style & Naming Conventions

TypeScript ESM across packages.
Prettier enforces tabs (useTabs: true, tabWidth: 2), semicolons, single quotes, and printWidth: 120.
ESLint rules are strict in packages/mcp and server code (no any, consistent type imports, prefer interface type definitions).
Tests use .test.ts or .spec.ts naming.

Testing Guidelines

Unit/integration tests run with Vitest in both packages/app and packages/mcp.
Run all tests with pnpm test or package-specific commands above.
No explicit coverage gate is configured; add tests for new tools, transports, or formatting logic.

Commit & Pull Request Guidelines

Recent history mixes merge commits and conventional-style messages like chore: release vX; no enforced convention.
Keep commits small and descriptive; add a scope when it clarifies the area (app, mcp).
PRs should include a concise summary, test results, linked issue (if applicable), and screenshots for UI changes under packages/app/src/web.

Read the full file on GitHub · 47 lines

Changes

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.

1mo ago Changed · -7 lines · -69 tokens per session 06ee4a54fe08
2mo ago First seen · 54 lines · 673 tokens per session scan A f8723bfac003

Subscribe to this mod's changes

hf-mcp-server AGENTS.md is an instructions file published in the GitHub repository huggingface/hf-mcp-server (310 stars, last pushed yesterday), licensed MIT. It adds 604 tokens to every session, about $0.0024 per session on Opus 5.5. 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.

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).

microsoft/vscode · 9,715 tokens

hf-mcp-server: Instructions file for Codex