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-peptide-target-deorphanizationgit 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-peptide-target-deorphanization)<a href="https://agentmods.dev/skills/mims-harvard/tooluniverse/tooluniverse-peptide-target-deorphanization"><img src="https://agentmods.dev/badge/skills/mims-harvard/tooluniverse/tooluniverse-peptide-target-deorphanization/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-peptide-target-deorphanization"><img src="https://agentmods.dev/badge/skills/mims-harvard/tooluniverse/tooluniverse-peptide-target-deorphanization.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to medium
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 →
- medium analysis-evasion · line 1 Suspicious Unicode normalization or mixed-script contentFix: Review the flagged content for security risks. Ensure no credentials, secrets, or sensitive data are exposed.
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 | $0.00195 | $0.04520 |
| Opus 5.5 | $0.00078 | $0.01808 |
| Sonnet 5.5 | $0.00039 | $0.00904 |
| Haiku 4.5 | $0.00019 | $0.00452 |
Grade A, and why
tooluniverse-peptide-target-deorphanization 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 — 130 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Peptide Target Deorphanization
Deorphanize a peptide: given a peptide sequence plus an observed phenotype (and often a hypothesized target the peptide does NOT actually bind), find its likely real protein target(s) — using ToolUniverse's keyless characterization, homology, target-family, phenotype, cross-species, and (optional, key-gated) co-folding tools.
The target can be anything, not just a GPCR. A target-class router (GPCR ligand / ion-channel toxin / protease target / cytokine or growth-factor receptor / integrin ligand / antimicrobial / unknown) classifies the peptide up front and adapts the enumeration strategy. GPCRs are the best-trodden case (and the validated control), but the pipeline's spine — homology, motif, phenotype, cross-species, co-fold — is target-class-agnostic, and family enumeration uses HGNC gene-family (general) + InterPro (general) + GPCRdb (GPCR-only cross-check).
Core reasoning: LOOK UP, DON'T GUESS
The failure mode this skill defends against is guessing a target from the peptide's name or its assumed mechanism. A peptide can be phenotypically active yet not bind the hypothesized target — because it hits a paralog, a different family member, or the same receptor in a different species whose binding interface has diverged. So:
- Never assert a target from memory. Every candidate must come from a tool result (homology hit, family enumeration, phenotype association, or structural co-fold), with the tool name and accession recorded.
- Anchor on PHENOTYPE × STRUCTURE/SEQUENCE plausibility, not on the peptide's reputed mechanism. The real target is the intersection of (a) what the sequence/motif/structure says it could bind and (b) what the phenotype says is biologically relevant. A name-level guess ("it's a GLP-1 analog so it's GLP1R") is exactly what produces off-target errors.
- Reconcile across species. "Binds in species A but not B" is usually interface sequence divergence, not a different target. Always pull the ortholog set and align the candidate receptor's ligand-binding interface across the assay species before concluding the peptide "doesn't work."
- A non-binding result against the hypothesized target is a clue, not a dead end. It promotes the paralogs and phenotype-shared receptors to the top of the candidate list.
What ships with it
3 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.
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 Changed · +20 lines 7f4c6cf7d7c9
- 1mo ago First seen · 110 lines · 195 tokens per session scan A ac613ea17526
tooluniverse-peptide-target-deorphanization is a skill published in the GitHub repository mims-harvard/ToolUniverse (1,744 stars, last pushed yesterday), licensed Apache-2.0. It adds 195 tokens to every session and 4,520 once invoked, about $0.0008 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 2026-09-03.
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.