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/huggingface/ml-intern/agents-mdgit clone --depth 1 https://github.com/huggingface/ml-internWhat 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 | $0.01007 | $0.01007 |
| Opus 5 | $0.00504 | $0.00504 |
| Sonnet 5 | $0.00201 | $0.00201 |
| Haiku 4.5 | $0.00101 | $0.00101 |
Grade A, and why
ml-intern AGENTS.md scanned grade A with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
- Backend health check: `curl -g http://[::1]:7860/api` How it starts
The opening of the file, as written. The whole thing — 58 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent Notes
Local Dev Servers
- Frontend: from
frontend/, runnpm ciif dependencies are missing, thennpm run dev. - Backend: from
backend/, runuv run uvicorn main:app --host ::1 --port 7860. - Frontend URL: http://localhost:5173/
- Backend health check:
curl -g http://[::1]:7860/api - Frontend proxy health check:
curl http://localhost:5173/api
Notes:
- Vite proxies
/apiand/authtohttp://localhost:7860. - If
127.0.0.1:7860is already owned by another local process, binding the backend to::1lets the Vite proxy resolvelocalhostcleanly. - Prefer
npm ciovernpm installfor setup, sincenpm installmay rewritefrontend/package-lock.jsonmetadata depending on npm version. - Non-local LLM calls use
https://router.huggingface.co/v1with the active Hugging Face user's token. Web sessions and the CLI default to GLM 5.2. For local development, setHF_TOKENand optionallyML_INTERN_DEFAULT_MODEL_ID. - When asked to start the local server, export the GitHub CLI token first with
export GITHUB_TOKEN="$(gh auth token)". - When debugging a web app issue tied to a session ID, inspect the session data in
smolagents/ml-intern-sessionsfor additional context.
Development Checks
- Before every commit, run
uv run ruff check .anduv run ruff format --check .. - If formatting fails, run
uv run ruff format ., then re-run the Ruff checks before committing.
Git Workflow
- Before creating any new branch or worktree, switch to
mainand pull the latest changes.
GitHub CLI
- Always use the
ghCLI for GitHub operations such as opening, editing, inspecting, or commenting on PRs and issues. - For multiline PR descriptions, prefer
gh pr edit <number> --body-file <file>over inline--bodyso shell quoting,$env-var names, backticks, and newlines are preserved correctly. - If
ghreports an invalid token or auth failure, retry the command withGH_TOKENandGITHUB_TOKENunset, for exampleenv -u GH_TOKEN -u GITHUB_TOKEN gh pr create ..., soghcan use the stored login token instead of a stale environment token. - In Codex, sandboxed
ghauth checks can report a valid keyring login as invalid when GitHub network access is restricted. Before telling the user to re-authenticate, retry with both env tokens unset and GitHub network access enabled.
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.
- 2d ago First seen · 58 lines · 1,007 tokens per session scan A 2252def186a8
ml-intern AGENTS.md is an instructions file published in the GitHub repository huggingface/ml-intern (10,759 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 1,007 tokens to every session, about $0.0050 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
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.