devtu-self-evolve

Orchestrate the ToolUniverse self-improvement cycle from discovery to pull requests.

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

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Coordinates and automates the end-to-end developer loop for ToolUniverse to reduce manual coordination, eliminate noisy agent false positives, and ensure verified, high-quality tool updates are produced and landed safely.

Core Features & Use Cases

  • Phase orchestration: Runs discovery, tool creation, researcher-persona testing, verified fixes, optimization, documentation checks, and shipping in sequence.
  • Persona-driven testing: Launches paired researcher agents with scenario templates to surface real-world failures and severity-tagged issues.
  • Verification-first fixes: Enforces CLI verification before code changes, runs a code-simplifier, linters, and guided commit/PR workflows to prevent regressions.
  • Release discipline: Automates branch creation, rebase rules, push policies, and PR validation to comply with Git rules and repository conventions.
  • Use Case: Run a test round to discover API gaps, build or repair a tool, validate fixes with persona agents, optimize descriptions and skills, and open a PR for review.

Quick Start

Use the devtu-self-evolve skill to run a full ToolUniverse test round with researcher personas, verify reported issues via the CLI, apply validated fixes, run the code-simplifier and linters, and prepare a branch for a pull request.

Frequently Asked Questions about devtu-self-evolve

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

FAQPage Schema
How do I automate the ToolUniverse self-improvement cycle for API discovery and testing?

Enforce verification-first fixes by requiring CLI verification of reported issues before applying code changes. The workflow then runs a code-simplifier and linters to prevent regressions and ensure validated, high-quality tool updates.

How does persona-driven testing work for life-science tool development?

Persona-driven testing launches paired researcher agents with scenario templates to surface real-world failures during tool development. This mechanism generates severity-tagged issues to ensure tools are validated against realistic usage patterns.

What is the best way to ship verified tool updates via GitHub pull requests?

Ship verified tool updates by automating branch creation, rebase rules, push policies, and PR validation. This release discipline complies with Git rules and repository conventions to land changes safely via pull requests.

Does the ToolUniverse self-improvement workflow require manual Git branch management?

The ToolUniverse self-improvement workflow does not require manual Git branch management. It automates branch creation, rebase rules, and PR validation to comply with repository conventions, preparing changes for GitHub safely.

Why do my agent testing workflows produce noisy false positives in tool development?

Agent testing workflows produce noisy false positives when lacking coordinated verification. This orchestration skill enforces CLI-based test reproduction and bug diagnosis before fixes, eliminating false positives and ensuring high-quality updates.