devtu-optimize-skills

Optimize ToolUniverse research skills to produce evidence-graded narrative reports with source attribution.

Updated Apr 18, 2026
One-click install
npx skills add https://github.com/Centaurioun/osteogenesis_imperfecta --skill devtu-optimize-skills-centaurioun
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: devtu-optimize-skills
Source: https://github.com/Centaurioun/osteogenesis_imperfecta/tree/main/.agents/skills/devtu-optimize-skills
Command: npx skills add https://github.com/Centaurioun/osteogenesis_imperfecta --skill devtu-optimize-skills-centaurioun

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill provides a structured, test-driven approach to fixing and improving ToolUniverse research skills so they produce reliable, evidence-graded, and reproducible reports instead of noisy or non-functional outputs.

Core Features & Use Cases

  • Tool Verification: Check and correct tool parameter contracts before any calls to avoid runtime failures.
  • Disambiguation-First Workflow: Resolve identifiers and detect naming collisions before broad searches to reduce noise.
  • Evidence Grading & Completeness: Apply T1-T4 evidence tiers, quantified minimums, aggregated data gaps, and a mandatory completeness checklist.
  • Testing & Fallbacks: Require comprehensive real-data tests, fallback chains, and documented failure handling to ensure production readiness.
  • Use Case: Review an existing ToolUniverse research skill that returns inconsistent literature results and transform it into a validated, evidence-layered report with a JSON bibliography and a methods appendix.

Quick Start

Run a tool verification pass, disambiguate identifiers, and synthesize an evidence-graded narrative report with sources and a completeness checklist.

Frequently Asked Questions about devtu-optimize-skills

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I optimize ToolUniverse research skills to produce evidence-graded reports?

To optimize ToolUniverse research skills for evidence-graded reports, apply a test-driven approach that verifies tool parameter contracts, executes disambiguation-before-search workflows, and applies T1-T4 evidence grading with quantified completeness checks.

What is the disambiguation-first workflow for research skill identifier resolution?

The disambiguation-first workflow resolves identifiers and detects naming collisions before executing broad aggregator-first queries, reducing noisy results and ensuring accurate source attribution in narrative reports.

How do I apply T1-T4 evidence grading and completeness checks to research outputs?

Apply T1-T4 evidence grading by categorizing aggregated data into tiered evidence levels, calculating quantified minimums, identifying data gaps, and validating the final output against a mandatory completeness checklist.

Why do my research skills return inconsistent literature results and fail at runtime?

Inconsistent literature results and runtime failures occur when tool parameter contracts are unverified or fallback handling is missing, requiring comprehensive real-data testing and documented failure chains for production readiness.

What is the best way to structure fallback handling for aggregator-first queries?

The best way to structure fallback handling for aggregator-first queries is to establish comprehensive real-data tests, define explicit fallback chains for tool failures, and document failure handling to ensure reproducible report synthesis.