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-hla-immunogenomicsgit 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-hla-immunogenomics)<a href="https://agentmods.dev/skills/mims-harvard/tooluniverse/tooluniverse-hla-immunogenomics"><img src="https://agentmods.dev/badge/skills/mims-harvard/tooluniverse/tooluniverse-hla-immunogenomics/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-hla-immunogenomics"><img src="https://agentmods.dev/badge/skills/mims-harvard/tooluniverse/tooluniverse-hla-immunogenomics.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00100 | $0.04086 |
| Opus 5.5 | $0.00040 | $0.01634 |
| Sonnet 5.5 | $0.00020 | $0.00817 |
| Haiku 4.5 | $0.00010 | $0.00409 |
Grade A, and why
tooluniverse-hla-immunogenomics 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 — 258 lines — stays where its author put it; the contents beside it link to each section on GitHub.
HLA & Immunogenomics Analysis
Pipeline for exploring HLA gene families, MHC-peptide binding, epitope associations, and their clinical implications in transplantation, vaccine development, and cancer immunotherapy. Bridges immunogenetic databases (IMGT, IEDB) with functional annotation (UniProt) and druggability data (DGIdb).
Reasoning Strategy
HLA analysis is fundamentally about peptide presentation: the polymorphism of HLA molecules determines which peptides are displayed to T cells, which in turn governs disease susceptibility, transplant rejection, drug hypersensitivity, and vaccine immunogenicity. HLA type affects disease susceptibility for autoimmune conditions (HLA-B27 and ankylosing spondylitis), transplant rejection (HLA mismatch drives alloresponse), drug hypersensitivity (abacavir causes severe hypersensitivity reactions only in HLA-B*57:01 carriers), and vaccine design (epitopes must be presented by the recipient's HLA alleles to elicit a T-cell response). Class I and Class II HLA molecules have fundamentally different binding grooves, peptide lengths, and T-cell partners — never conflate them. The absence of an epitope from IEDB means it has not been tested, not that it cannot bind.
LOOK UP DON'T GUESS: Never assume an allele's binding properties or population frequency — query IEDB for experimental binding data and IMGT for allele annotation. Do not guess which HLA alleles are common in a population; look up published frequency data via PubMed.
Guiding principles:
- HLA nomenclature precision -- HLA allele names follow strict conventions (e.g., HLA-A*02:01); get the resolution level right
- MHC class awareness -- Class I (A, B, C) and Class II (DR, DQ, DP) have different binding properties and clinical roles
- Species context -- most queries target human HLA, but MHC exists across vertebrates; confirm species early
- Evidence layering -- combine binding data (IEDB) with gene annotation (IMGT) and structural context (UniProt)
- Clinical translation -- connect molecular findings to transplant matching, vaccine targets, or immunotherapy response
- English-first queries -- use English terms in all tool calls; respond in the user's language
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 · +25 lines 9272e805c454
- 1mo ago First seen · 233 lines · 100 tokens per session scan A 3adbe45f8a41
tooluniverse-hla-immunogenomics is a skill published in the GitHub repository mims-harvard/ToolUniverse (1,744 stars, last pushed yesterday), licensed Apache-2.0. It adds 100 tokens to every session and 4,086 once invoked, about $0.0004 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.