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-share-computergit 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-share-computer)<a href="https://agentmods.dev/skills/mims-harvard/tooluniverse/tooluniverse-share-computer"><img src="https://agentmods.dev/badge/skills/mims-harvard/tooluniverse/tooluniverse-share-computer/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-share-computer"><img src="https://agentmods.dev/badge/skills/mims-harvard/tooluniverse/tooluniverse-share-computer.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 | $0.00123 | $0.03152 |
| Opus 5.5 | $0.00049 | $0.01261 |
| Sonnet 5.5 | $0.00025 | $0.00630 |
| Haiku 4.5 | $0.00012 | $0.00315 |
Grade C, and why
tooluniverse-share-computer scanned grade C with 3 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Downloads and executes remote codehighSupply chain
curl | sh runs whatever the server returns today, which is not necessarily what it returned when this was reviewed.
curl -LsSf https://connect.aiscientist.tools/install.sh | sh Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
curl -LsSf https://connect.aiscientist.tools/install.sh | sh Runs shell commandslowCapability
Expected in a hook, worth knowing in a rule or an instructions file.
"$PY" -c 'import subprocess, sys; log = open(sys.argv[1], "w"); p = subprocess.Popen(sys.argv[2:], stdin=subprocess.DEVNULL, stdout=log, stderr=subprocess.STDOUT, start_new_session=True); print(p.pid)' \ How it starts
The opening of the file, as written. The whole thing — 184 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Share a computer's tools through ToolUniverse
The person you are helping is usually a scientist, not a programmer. Do every technical step yourself. Ask them only what you cannot find out, in plain words, one question at a time. Never ask them to edit code, read logs, or type commands; never paste or ask for keys, tokens or passwords in the chat.
Nothing leaves their computer except what the chosen tool returns. ToolUniverse only relays calls to it; their files, data and passwords stay where they are, and only the tools they pick are reachable.
0. Check you can do this here
You must be able to run commands on the computer that has the code or model (Claude Code, Codex, or another agent with a terminal there). If you cannot (for example you are Claude in a browser), say so plainly and stop:
Sharing has to be set up on the computer that has your model. Open Claude Code or Codex on that computer and ask it the same thing; it can use this guide too.
Sandboxes. Several steps need the internet and write to the user's home folder (installing, signing in, sharing). If your environment runs commands in a sandbox (Codex does by default), a step can fail with errors such as Could not resolve host or Permission denied. That is your own safety setting, not a broken installer: ask the user to approve running the step with network access, in one plain sentence ("I need internet access to install ToolUniverse's sharing tool. Allow it?"). Combine steps into as few commands as you can, so the user approves as few times as possible.
1. Find out what to share (ask, don't assume)
Work out, from the conversation or by looking around with their permission:
- What: a function or script they wrote, a notebook cell, a model they run, a program that already serves MCP, or one of ToolUniverse's reviewed models (run
tu remote listafter step 2 to see them: ESM, Boltz, scVI, Enformer, and others). - The one thing it should do, as a sentence ("score a protein sequence", "look up a sample in our lab database"). Share narrow operations, not a whole program.
- Inputs and output: what a caller gives and gets back.
- Name: what colleagues should see, such as "Smith Lab GPU".
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 Changed bfd8361cca14
- 7d ago Changed · +29 lines a771b164ded1
- 8d ago First seen · 155 lines · 123 tokens per session scan C 0e1eec70c28f
tooluniverse-share-computer is a skill published in the GitHub repository mims-harvard/ToolUniverse (1,744 stars, last pushed yesterday), licensed Apache-2.0. It adds 123 tokens to every session and 3,152 once invoked, about $0.0005 per session on Opus 5.5. A static security scan graded it C with 3 findings (downloads and executes remote code, makes network calls, runs shell commands). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-10-09.
Other skills, from other repositories
ai-agent-tool-routing
Use when designing or reviewing an AI agent tool registry or MCP router: document parameter schemas, caller authorization, execution limits, replay-safe retries and verification evidence.
nanoresearch-writing
Draft a LaTeX research paper from all previous stage outputs.
construct-toy-examples
Generate and analyze simpler examples that satisfy both the assumptions and the conclusion of a theorem statement or subgoal. Use when you are stuck in reasoning and need simpler examples to regain traction, or when you want to see where the assumptions take effect and gain intuition.
obtain-immediate-conclusions
Derive immediate mathematical consequences from a theorem statement or subgoal. Use when starting a new problem, branch, or subgoal, or when cheap progress or a cleaner reformulation is needed before deeper proof search.
clinical-research-analysis-framework
Guided workflow for statistical and ML analysis of clinical data. Use when planning or executing research analyses on MIMIC, eICU, or similar EHR data. Ensures methodological rigor through structured consultation, assumption checking, and stepwise execution with audit trails.
apache-iv-score
Calculate APACHE IV (Acute Physiology and Chronic Health Evaluation IV) score for ICU mortality prediction. Use for severity assessment, hospital mortality prediction, ICU benchmarking, or case-mix adjustment. eICU has pre-computed scores; MIMIC-IV requires custom implementation with diagnosis mapping challenges.