ToolUniverse is a collection of tools, interfaces, and supporting components for building AI systems that perform scientific work. It is for developers creating AI scientist agents that use APIs, databases, machine-learning tools, and domain-specific utilities. The catalogue includes skills, commands, an MCP server, an agent, and a hook for working with the ecosystem.
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 skills add mims-harvard/ToolUniverse --skill tooluniverse-ml-inference-utilitygit clone --depth 1 https://github.com/mims-harvard/ToolUniverseWrote 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/skills/mims-harvard/tooluniverse/tooluniverse-ml-inference-utility)<a href="https://agentmods.dev/skills/mims-harvard/tooluniverse/tooluniverse-ml-inference-utility"><img src="https://agentmods.dev/badge/skills/mims-harvard/tooluniverse/tooluniverse-ml-inference-utility/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.
<a href="https://agentmods.dev/skills/mims-harvard/tooluniverse/tooluniverse-ml-inference-utility"><img src="https://agentmods.dev/badge/skills/mims-harvard/tooluniverse/tooluniverse-ml-inference-utility.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.00000 | $0.03860 |
| Opus 5.5 | $0.00000 | $0.01544 |
| Sonnet 5.5 | $0.00000 | $0.00772 |
| Haiku 4.5 | $0.00000 | $0.00386 |
Grade A, and why
tooluniverse-ml-inference-utility 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 10d 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 — 235 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Generic ML/NLP/Vision Inference (HuggingFace)
Run any HuggingFace-hosted model for a standard inference task via ToolUniverse's
10 HFInference_* tools. This is a utility skill: reach for it as a supporting
step inside another workflow (e.g. summarizing a batch of retrieved abstracts,
extracting entities from clinical text, embedding sentences for a similarity
search) rather than treating it as a research domain in its own right.
Honesty contract (read first) — the token requirement is not actually optional
Every tool's own JSON description says HF_TOKEN is an optional API key that
merely "raises rate limits." Live testing on this date contradicts that: all
10 tools, tested against the exact model IDs in their own test_examples, fail
with the identical error with no token set:
$ tu test HFInference_classify_text
✗ example 1 tool returned error: Unauthorized for
distilbert/distilbert-base-uncased-finetuned-sst-2-english. The model may be
gated/private; set a valid HF_TOKEN with access.
This reproduces across text and vision tools alike, and a dummy/invalid token
produces the exact same Unauthorized message (confirmed live) — so this is a
"needs a real, valid HuggingFace access token" requirement, not a cosmetic rate
limit. Treat HF_TOKEN as required in practice even though the tool schema
calls it optional:
- Check for
HF_TOKENbefore promising a result. If it's unset, tell the user plainly that HuggingFace's serverless inference now requires a token, point them to https://huggingface.co/settings/tokens to create a free one (aread-scope token is sufficient), and stop rather than guessing an output. - Never fabricate a classification, embedding, summary, or entity list. If the call fails, report the failure — do not describe what a model would probably say.
- A working call can still fail per-model. Some models are gated (require
accepting a license on the HuggingFace model page even with a valid token)
or cold (first call after inactivity can 503 — retry once with
wait_for_model: truebefore concluding a model is broken).
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 10d ago First seen · 235 lines · 0 tokens per session scan A ae4213917c21
tooluniverse-ml-inference-utility is a skill published in the GitHub repository mims-harvard/ToolUniverse (1,744 stars, last pushed yesterday), licensed Apache-2.0. It costs nothing until one of its globs matches a file; then it loads 3,860 tokens. 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 2026-10-07.
Other skills, from other repositories
external-repo-setup
Protocol for GPU agents to clone external GitHub repos, install their dependencies, download pretrained weights, and integrate them as featurizers or models inside a focus-area train.py. Use this when a proposal references a GitHub repo or pretrained checkpoint that is not already present in the focus-area workspace.
hugging-face-datasets
Create and manage datasets on Hugging Face Hub. Supports initializing repos, defining configs/system prompts, streaming row updates, and SQL-based dataset querying/transformation. Designed to work alongside HF MCP server for comprehensive dataset workflows.
thinking-out-loud
A contract for what the agent does when a long, messy, stream-of-consciousness ramble arrives (usually voice dictation): act on nothing until the echo brief is approved. The echo audits the entire transfer, mission, locked decisions and constraints, open questions, flips and parked tangents, with the model's…
pydantic-ai
Build production-ready AI agents with PydanticAI — type-safe tool use, structured outputs, dependency injection, and multi-model support.
capacity
Discovers available Azure OpenAI model capacity across regions and projects. Analyzes quota limits, compares availability, and recommends optimal deployment locations based on capacity requirements. USE FOR: find capacity, check quota, where can I deploy, capacity discovery, best region for capacity, multi-project…
airunway-aks-setup
Set up AI Runway on AKS — from bare cluster to running model. Covers cluster verification, controller install, GPU assessment, provider setup, and first deployment. WHEN: "setup AI Runway", "onboard AKS cluster", "install AI Runway", "airunway setup", "deploy model to AKS", "GPU inference on AKS", "KAITO setup on…