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 setup-tooluniversegit 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/setup-tooluniverse)<a href="https://agentmods.dev/skills/mims-harvard/tooluniverse/setup-tooluniverse"><img src="https://agentmods.dev/badge/skills/mims-harvard/tooluniverse/setup-tooluniverse/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/setup-tooluniverse"><img src="https://agentmods.dev/badge/skills/mims-harvard/tooluniverse/setup-tooluniverse.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- Socket pass
- Snyk fail
- NVIDIA SkillSpector warn
SkillSpector: 18 findings, up to high
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- high Tool Misuse · line 51 Tool calls are chained to bypass individual safety checks or escalate capabilities beyond what any single tool call would allow.Fix: Limit tool chaining depth and validate the output of each tool before passing it to the next. Require explicit user approval for multi-step chains.
- high Agent Snooping · line 224 Skill reads from agent configuration directories (.claude/, .codex/, .gemini/). These directories may contain API keys, personal settings, and other credentials that the skill has no legitimate need to access.Fix: Remove all code or instructions that access agent configuration directories (.claude/, .codex/, .gemini/). If configuration values are needed, pass them explicitly as parameters or environment variabl
- high Tool Misuse · line 313 Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).Fix: Validate all tool parameters against an allowlist. Reject dangerous parameter values (shell=True, --force, -rf /) and use safe defaults.
- high Tool Misuse · line 313 Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).Fix: Validate all tool parameters against an allowlist. Reject dangerous parameter values (shell=True, --force, -rf /) and use safe defaults.
- high Tool Misuse · line 313 Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).Fix: Validate all tool parameters against an allowlist. Reject dangerous parameter values (shell=True, --force, -rf /) and use safe defaults.
- medium MCP Rug Pull · line 3 uvx/uv tool run commands without ==version create a rug-pull risk.Fix: Pin the version: uvx package-name==1.2.3
- medium MCP Rug Pull · line 62 uvx/uv tool run commands without ==version create a rug-pull risk.Fix: Pin the version: uvx package-name==1.2.3
- medium MCP Rug Pull · line 63 uvx/uv tool run commands without ==version create a rug-pull risk.Fix: Pin the version: uvx package-name==1.2.3
- medium MCP Rug Pull · line 64 uvx/uv tool run commands without ==version create a rug-pull risk.Fix: Pin the version: uvx package-name==1.2.3
- medium MCP Rug Pull · line 65 uvx/uv tool run commands without ==version create a rug-pull risk.Fix: Pin the version: uvx package-name==1.2.3
- medium MCP Rug Pull · line 82 uvx/uv tool run commands without ==version create a rug-pull risk.Fix: Pin the version: uvx package-name==1.2.3
- medium MCP Rug Pull · line 416 uvx/uv tool run commands without ==version create a rug-pull risk.Fix: Pin the version: uvx package-name==1.2.3
- medium MCP Rug Pull · line 419 uvx/uv tool run commands without ==version create a rug-pull risk.Fix: Pin the version: uvx package-name==1.2.3
- medium MCP Rug Pull · line 435 uvx/uv tool run commands without ==version create a rug-pull risk.Fix: Pin the version: uvx package-name==1.2.3
- medium Agent Snooping · line 308 Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.Fix: Remove all code or instructions that list or read other skills' files or directories. Skills should operate independently; cross-skill access is a privilege escalation.
- medium Agent Snooping · line 311 Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.Fix: Remove all code or instructions that list or read other skills' files or directories. Skills should operate independently; cross-skill access is a privilege escalation.
- medium Privilege Escalation · line 411 Commands invoke sudo or root privileges. Verify this elevated access is necessary and justified.Fix: Avoid sudo/root unless strictly required. Prefer least-privilege patterns. If elevation is needed, document the justification and scope.
- low Supply Chain · line 51 Remote code is downloaded and executed. This bypasses code review and could introduce malicious code.Fix: Avoid downloading and executing remote scripts. Use trusted packages from PyPI/npm. If remote fetch is required, verify checksums and use HTTPS.
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 | $0.00208 | $0.05563 |
| Opus 5 | $0.00104 | $0.02781 |
| Sonnet 5 | $0.00042 | $0.01113 |
| Haiku 4.5 | $0.00021 | $0.00556 |
Grade A, and why
setup-tooluniverse 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 — 439 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Setup ToolUniverse
Guide the user step-by-step through setting up ToolUniverse.
Agent Behavior
- Detect language from user's first message. Respond in their language; keep commands/URLs in English.
- Go one step at a time. Ask before proceeding.
- Use AskQuestion for structured choices.
- Explain briefly in plain language. Celebrate small wins.
- When something goes wrong, help troubleshoot before moving on.
Internal Notes (do not show)
ToolUniverse has 1200+ tools. The tooluniverse command enables compact mode automatically, exposing only 5 core MCP tools (list_tools, grep_tools, get_tool_info, execute_tool, find_tools) while keeping all tools accessible via execute_tool.
What is ToolUniverse?
Always explain first, in plain language:
ToolUniverse is free, open-source software connecting to 2,000+ scientific databases (PubMed, UniProt, ChEMBL, FAERS, ClinicalTrials.gov, etc.). Instead of visiting each website, you search from one place. Think of it like a universal remote for scientific databases.
Why AI assistants? The AI reads your question, figures out which databases to search, runs queries, and summarizes results. You just ask your question.
Step 1: Choose How to Use It
Present using AskQuestion:
| Mode | What it means | Who it's for |
|---|---|---|
| Chat mode | Ask questions to an AI assistant. No coding. | Most researchers. |
| Command line | Type short commands in Terminal. | Quick tests. Terminal-comfortable users. |
| Python code | Write scripts for automated pipelines. | Programmers. |
Options: "I want to ask questions" → Chat mode | "Quick try" → CLI | "I write Python" → SDK | "I don't know" → Recommend Chat mode
If Chat mode, ask which app (AskQuestion): Cursor, Claude Desktop, VS Code/Copilot, Windsurf, Claude Code, Gemini CLI, Codex, Cline/Trae/Antigravity/OpenCode. "I don't have any" → Recommend Claude Desktop.
Step 2: Install uv
What ships with it
14 files 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.
- API_KEYS_REFERENCE.md 14 KB
- CLAUDE_DESKTOP.md 6.2 KB
- EXAMPLES.md 9.2 KB
- GITHUB_ISSUE.md 2.3 KB
- INSTALL.md 3.6 KB
- MCP_CONFIG.md 11 KB
- references/mcp-configs.md 1.2 KB
- scripts/check_prerequisites.py 5.3 KB runs code
- scripts/diagnose_setup.py 11 KB runs code
- scripts/list_tool_categories.py 2.7 KB runs code
- scripts/verify_installation.py 7.5 KB runs code
- SKILLS_CATALOG.md 5.9 KB
- TRAE.md 7.2 KB
- TROUBLESHOOTING.md 7.5 KB
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 · +11 lines c7af36a024a1
- 9d ago First seen · 428 lines · 208 tokens per session scan F 6d802b7edf4a
setup-tooluniverse is a skill published in the GitHub repository mims-harvard/ToolUniverse (1,676 stars, last pushed today), licensed Apache-2.0. It adds 208 tokens to every session and 5,563 once invoked, about $0.0010 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 2026-08-30.
Other skills, from other repositories
dfam-check
Measure mesh files against Design for Additive Manufacturing (DfAM) rules and report printability findings per process (FDM, SLS, SLA/DLP, metal PBF, MJF). Use when the user asks whether a part is printable, wants overhang/wall-thickness/support analysis of an .stl, .obj, .ply, or .3mf mesh, wants a build-orientation…
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.
intermediate-outputs
Use this skill when working with circuit discovery in language models, mechanistic interpretability, activation patching, attribution patching, or Layer-wise Relevance Propagation (LRP) for neural network analysis.
clip-dissect
Use this skill when you need to automatically describe or interpret the functionality of individual neurons in deep neural networks (DNNs) using CLIP-based semantic analysis, perform mechanistic interpretability research on vision models, dissect convolutional or transformer-based image classifiers, identify what…