CtrlK
BlogDocsLog inGet started
Tessl Logo

tooluniverse-sdk

Build AI scientist systems with the ToolUniverse Python SDK for scientific research. Covers the 3 calling patterns (`tu.run` portable dict API, `tu.tools.X` function API, direct class instantiation), tool loading, batch execution, MCP server integration, and embedding-based tool search. Use for SDK programming, custom tool composition, benchmarking pipelines, and integrating ToolUniverse into research workflows.

68

Quality

82%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Low

Low-risk findings worth noting

The canonical home for this skill is tooluniverse-sdk in mims-harvard/ToolUniverse

SKILL.md
Quality
Evals
Security

Quality

Content

72%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

Highly actionable, executable content with tight code examples and a clear core sequence. Two weaknesses hold it back: batch workflows lack result-validation checkpoints, and the lone progressive-disclosure reference points to a file that does not exist in the bundle.

Suggestions

Add a validation/verification step in batch workflows — e.g., check run_batch results for errors per-call before downstream processing — to lift workflow clarity above the batch-validation cap.

Create the referenced references/REFERENCE.md (or remove the dangling REFERENCE.md link) so progressive disclosure points to a real file.

Verify intermediate results in the scientific workflow (e.g., guard targets['data'] access after the OpenTargets call) to add an explicit checkpoint.

DimensionReasoningScore

Conciseness

Lean and code-dense with no basic-concept padding; every section earns its place. Minor inline over-explanation (e.g., '# REQUIRED before any tool call') could be trimmed slightly.

4 / 5

Actionability

Copy-paste-ready, fully executable code across Quick Start, Batch Execution, Scientific Workflow, Configuration, and Error Handling, covering the common cases concretely.

5 / 5

Workflow Clarity

Has a clear install→load_tools→find→execute sequence and a Critical Notes checklist, but the batch execution and scientific workflow examples lack validation of intermediate results before proceeding. Per the rubric, batch operations without validation cap workflow clarity at 3.

3 / 5

Progressive Disclosure

Section structure is reasonable and most content is well placed, but the body references REFERENCE.md ('See [REFERENCE.md](REFERENCE.md) for detailed guides') and no such file or references/ directory exists — a broken one-level-deep reference that undermines navigation.

3 / 5

Total

15

/

20

Passed

Description

92%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong, specific description that names the SDK, enumerates concrete capabilities across multiple calling patterns, and gives an explicit 'Use for' trigger clause. Slight room to add more natural synonyms in the trigger phrasing.

DimensionReasoningScore

Specificity

Lists multiple concrete actions — 'tool loading', 'batch execution', 'MCP server integration', 'embedding-based tool search', 'custom tool composition', 'benchmarking pipelines' — giving comprehensive coverage of the SDK's capabilities.

5 / 5

Completeness

Clearly states what the skill does (the 3 calling patterns plus listed features) and provides an explicit 'Use for ...' trigger clause with concrete phrases, answering both what and when.

5 / 5

Trigger Term Quality

Includes natural domain phrases users would say ('SDK programming', 'custom tool composition', 'benchmarking pipelines', 'integrating ToolUniverse into research workflows'); good coverage but a few natural synonyms and concrete trigger variations are missing.

4 / 5

Distinctiveness Conflict Risk

Tied to a named, niche SDK ('ToolUniverse Python SDK for scientific research') with distinct triggers, giving minimal conflict risk with other skills.

5 / 5

Total

19

/

20

Passed

Validation

93%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

relative_links

Relative link issues: 1 missing

Warning

Total

15

/

16

Passed

Repository
mims-harvard/ToolUniverse
Reviewed

Table of Contents

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.