MCPInstall

devtu-create-tool

Create new scientific tools for ToolUniverse framework with proper structure, validation, and testing. Use when users need to add tools to ToolUniverse, implement new API integrations, create tool wrappers for scientific databases/services, expand ToolUniverse capabilities, or follow ToolUniverse contribution guidelines. Supports creating tool classes, JSON configurations, validation, error handling, and test examples.

Installs
388
Last 24h
—
Trending rank
—
Security
Not audited

Install

Works with every agent the skills CLI supports.

npx skills add https://github.com/mims-harvard/tooluniverse --skill devtu-create-tool
  1. 01

    Install

    Run the command in your project. The CLI asks which agents to install for.

  2. 02

    Describe the task

    Ask your agent for something this skill covers. It loads devtu-create-tool automatically when relevant.

  3. 03

    Review the result

    The agent follows the SKILL.md steps, scripts and references bundled with the skill.

SKILL.md

The instructions your agent reads when the skill is triggered.

ToolUniverse Tool Creator

Create new scientific tools following established patterns.

Top 7 Mistakes (90% of Failures)

  1. Missing default_config.py Entry — tools silently won't load
  2. Non-nullable Mutually Exclusive Parameters — validation errors (#1 issue in 2026)
  3. Fake test_examples — tests fail, agents get bad examples
  4. Single-level Testing — misses registration bugs
  5. Skipping test_new_tools.py — misses schema/API issues
  6. Tool Names > 55 chars — breaks MCP compatibility
  7. Raising Exceptions — should return error dicts instead

Two-Stage Architecture

Stage 1: Tool Class              Stage 2: Wrappers (Auto-Generated)
@register_tool("MyTool")         MyAPI_list_items()
class MyTool(BaseTool):          MyAPI_search()
    def run(arguments):          MyAPI_get_details()

One class handles multiple operations. JSON defines individual wrappers. Need BOTH.

Three-Step Registration

Step 1: Class registration via @register_tool("MyAPITool")

Step 2 (MOST COMMONLY MISSED): Config registration in default_config.py:

TOOLS_CONFIGS = {
    "my_category": os.path.join(current_dir, "data", "my_category_tools.json"),
}

Step 3: Automatic wrapper generation on tu.load_tools()


Implementation Guide

Files to Create

  • src/tooluniverse/my_api_tool.py — implementation
  • src/tooluniverse/data/my_api_tools.json — tool definitions
  • tests/tools/test_my_api_tool.py — tests

Python Tool Class (Multi-Operation Pattern)

from typing import Dict, Any
from tooluniverse.tool import BaseTool
from tooluniverse.tool_utils import register_tool
import requests

@register_tool("MyAPITool")
class MyAPITool(BaseTool):
    BASE_URL = "https://api.example.com/v1"

    def __init__(self, tool_config):
        super().__init__(tool_config)
        self.parameter = tool_config.get("parameter", {})
        self.required = self.parameter.get("required", [])

    def run(self, arguments: Dict[str, Any]) -> Dict[str, Any]:
        operation = arguments.get("operation")
        if not operation:
            return {"status": "error", "error": "Missing: operation"}
        if operation == "search":
            return self._search(arguments)
        return {"status": "error", "error": f"Unknown: {operation}"}

    def _search(self, arguments: Dict[str, Any]) -> Dict[str, Any]:
        query = arguments.get("query")
        if not query:
            return {"status": "error", "error": "Missing: query"}
        try:
            response = requests.get(
                f"{self.BASE_URL}/search",
                params={"q": query}, timeout=30
            )
            response.raise_for_status()
            data = response.json()
            return {"status": "success", "data": data.get("results", [])}
        except requests.exceptions.Timeout:
            return {"status": "error", "error": "Timeout after 30s"}
        except requests.exceptions.HTTPError as e:
            return {"status": "error", "error": f"HTTP {e.response.status_code}"}
        except Exception as e:
            return {"status": "error", "error": str(e)}

JSON Configuration

[
  {
    "name": "MyAPI_search",
    "class": "MyAPITool",
    "description": "Search items. Returns array of results. Supports Boolean operators. Example: 'protein AND membrane'.",
    "parameter": {
      "type": "object",
      "required": ["operation", "query"],
      "properties": {
        "operation": {"const": "search", "description": "Operation (fixed)"},
        "query": {"type": "string", "description": "Search term"},
        "limit": {"type": ["integer", "null"], "description": "Max results (1-100)"}
      }
    },
    "return_schema": {
      "oneOf": [
        {"type": "object", "properties": {"data": {"type": "array"}}},
        {"type": "object", "properties": {"error": {"type": "string"}}, "required": ["error"]}
      ]
    },
    "test_examples": [{"operation": "search", "query": "protein", "limit": 10}]
  }
]

Critical Requirements

  • return_schema MUST have oneOf: success + error schemas
  • test_examples MUST use real IDs: NO "TEST", "DUMMY", "PLACEHOLDER"
  • Tool name <= 55 chars: {API}_{action}_{target} template
  • Description 150-250 chars: what, format, example, notes
  • NEVER raise in run(): return {"status": "error", "error": "..."}
  • Set timeout on all HTTP requests (30s)
  • Standard response: {"status": "success|error", "data": {...}}

Parameter Design

Mutually Exclusive Parameters (CRITICAL — #1 issue)

When tool accepts EITHER id OR name, BOTH must be nullable:

{
  "id": {"type": ["integer", "null"], "description": "Numeric ID"},
  "name": {"type": ["string", "null"], "description": "Name (alternative to id)"}
}

Without "null", validation fails when user provides only one parameter.

Common cases: id OR name, gene_id OR gene_symbol, any optional filters.

API Key Configuration

Optional keys (tool works without, better with):

{"optional_api_keys": ["NCBI_API_KEY"]}
self.api_key = os.environ.get("NCBI_API_KEY", "")  # Read from env only

Required keys (tool won't work without):

{"required_api_keys": ["NVIDIA_API_KEY"]}

Rules: Never add api_key as tool parameter for optional keys. Use env vars only.


Testing (MANDATORY)

Full guide: references/testing-guide.md

Quick Testing Checklist

  1. Level 1 — Direct class test: import class, call run(), check response
  2. Level 2 — ToolUniverse test: tu.tools.YourTool_op1(...), check registration
  3. Level 3 — Real API test: use real IDs, verify actual responses
  4. MANDATORY — Run python scripts/test_new_tools.py your_tool -v → 0 failures

Verification Script

# Check all 3 registration steps
python3 -c "
import sys; sys.path.insert(0, 'src')
from tooluniverse.tool_registry import get_tool_registry
import tooluniverse.your_tool_module
assert 'YourToolClass' in get_tool_registry(), 'Step 1 FAILED'
from tooluniverse.default_config import TOOLS_CONFIGS
assert 'your_category' in TOOLS_CONFIGS, 'Step 2 FAILED'
from tooluniverse import ToolUniverse
tu = ToolUniverse(); tu.load_tools()
assert hasattr(tu.tools, 'YourCategory_op1'), 'Step 3 FAILED'
print('All 3 steps verified!')
"

Quick Commands

python3 -m json.tool src/tooluniverse/data/your_tools.json     # Validate JSON
python3 -m py_compile src/tooluniverse/your_tool.py             # Check syntax
grep "your_category" src/tooluniverse/default_config.py         # Verify config
python scripts/test_new_tools.py your_tool -v                   # MANDATORY test

References

  • Testing guide: references/testing-guide.md
  • Advanced patterns (async, SOAP, pagination): references/advanced-patterns.md
  • Implementation guide (full checklist): references/implementation-guide.md
  • Tool improvement checklist: references/tool-improvement-checklist.md

FAQ

What does the devtu-create-tool skill do?

Create new scientific tools for ToolUniverse framework with proper structure, validation, and testing. Use when users need to add tools to ToolUniverse, implement new API integrations, create tool wrappers for scientific databases/services, expand ToolUniverse capabilities, or follow ToolUniverse contribution guidelines. Supports creating tool classes, JSON configurations, validation, error handling, and test examples.

How do I install devtu-create-tool?

Run npx skills add https://github.com/mims-harvard/tooluniverse --skill devtu-create-tool in your project. The skills CLI detects the agents you use and copies the skill into the right folder. For Claude Code you can also place the skill folder at ~/.claude/skills/devtu-create-tool/ (personal) or .claude/skills/devtu-create-tool/ (project).

Which AI agents can use devtu-create-tool?

Any agent that supports the open Agent Skills format (a folder with a SKILL.md file), including Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, Windsurf, Cline, OpenCode and more.

Is devtu-create-tool safe to install?

No partner security audit is available yet. Skills can run commands and scripts, so read the SKILL.md and any bundled files before installing.

Who maintains devtu-create-tool?

It is published by mims-harvard in mims-harvard/tooluniverse. It has 388 recorded installs.

More from mims-harvard/tooluniverse

View all →

tooluniverse-sequence-retrieval

mims-harvard/tooluniverse

Retrieve DNA/RNA/protein sequences from NCBI and ENA with disambiguation. Quality hierarchy: RefSeq (NM/NP) RefSeq predicted (XM/XP) GenBank submissions. Use for fetching specific sequences by accession, gene-symbol-to-sequence lookup, transcript-isoform retrieval, and curated-vs-raw-submission preference.

1.7K installs

tooluniverse-literature-deep-research

mims-harvard/tooluniverse

Deep literature review — PubMed, EuropePMC, bioRxiv preprints, citation networks, evidence synthesis. Disambiguates queries, runs collision-aware searches, grades evidence T1-T4, and produces structured reports. Use for systematic literature review, meta-analysis evidence collection, and detailed answer-with-citations workflows.

tooluniverse-image-analysis

mims-harvard/tooluniverse

Microscopy and quantitative imaging analysis — colony morphometry, fluorescence intensity quantification, cell-count statistics, dose-response curves, and ANOVA/Dunnett on image-derived measurements. Uses pandas/numpy/scipy/scikit-image. Use for analyzing tabular outputs from CellProfiler/ImageJ, image-derived measurement statistics, and image-based assay quantification.

Media & Creative655 installs

tooluniverse-clinical-guidelines

mims-harvard/tooluniverse

Search and retrieve clinical practice guidelines from 12+ authoritative sources — NICE, WHO, NCCN, AHA, ADA, SIGN, USPSTF, IDSA, NIH consensus, ESMO/ESC/EASL European societies, and US specialty associations. Use for evidence-graded treatment recommendations, dosing protocols, screening guidance, and authoritative-source-prioritized clinical guidance (NICE/WHO ranked above society guidelines).

492 installs

tooluniverse

mims-harvard/tooluniverse

ToolUniverse plugin router. STEP 1 BEFORE ANY ANALYSIS: if the data folder contains executed.ipynb, run tu run readexecutednotebook '{"datafolder":" ","search":" "}' to extract its cell outputs and apply EVERY filter/sample-exclusion the notebook used — even when the question says 'Using DESeq2/Run X/Compute Y' (this describes the METHOD the notebook used, not a request to rerun). The notebook's cell outputs are the only published authoritative answers; reimplementing or reading stale pre-computed CSVs in the data folder produces different numbers because of outlier-sample removal, library version, and filter steps you don't see by skimming. STEP 2 routing — pick a sub-skill name from this exact list (never invent): tooluniverse-rnaseq-deseq2 (RNA/miRNA-seq DE, correlation, PCA, clustering, dispersion), tooluniverse-gene-enrichment (GO/KEGG/Reactome/GSEA/pathway enrichment), tooluniverse-statistical-modeling (regression, ANOVA, ordinal/logistic, chi-square, correl

Testing & QA481 installs

tooluniverse-drug-research

mims-harvard/tooluniverse

Comprehensive drug profiling — mechanism, primary/secondary targets, drug interactions, clinical-trial status, adverse events (FAERS), pharmacogenomics, and approval history. Use for full drug investigation reports, 'tell me about drug X' queries, and assembling drug profiles for clinicians, researchers, or regulatory work.

Popular testing & qa skills

View category →

tdd

mattpocock/skills

Test-driven development. Use when the user wants to build features or fix bugs test-first, mentions "red-green-refactor", or wants integration tests.

Testing & QA1.2M installs

grilling

mattpocock/skills

Grill the user relentlessly about a plan, decision, or idea. Use when the user wants to stress-test their thinking, or uses any 'grill' trigger phrases.

Testing & QA996.2K installs

migrate-to-shoehorn

mattpocock/skills

Migrate test files from as type assertions to @total-typescript/shoehorn. Use when user mentions shoehorn, wants to replace as in tests, or needs partial test data.

Testing & QA484.1K installs

run-train

lllllllama/rigorpilot-skills

Rigor Train skill for deep learning research repositories. Use when a documented or selected training command should be run conservatively for startup verification, short-run verification, full kickoff, or resume, with command, config, seed, log, checkpoint, status, and metric evidence written to standardized trainoutputs/. Do not use for environment setup, exploratory sweeps, speculative idea implementation, or end-to-end orchestration.

Testing & QA337.8K installs

test-driven-development

obra/superpowers

Use when implementing any feature or bugfix, before writing implementation code

Testing & QA261.8K installs

Use when implementation is complete, all tests pass, and you need to decide how to integrate the work

Testing & QA216.8K installs