agentforge-evolution

Diagnose AI agent faults and apply verifiable patches with safety gates.

Updated Apr 8, 2026
One-click install
npx skills add https://github.com/Kingxiao/agentforge --skill agentforge-evolution
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: agentforge-evolution
Source: https://github.com/Kingxiao/agentforge/tree/main/agentforge-evolution
Command: npx skills add https://github.com/Kingxiao/agentforge --skill agentforge-evolution

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

Diagnoses faults in AI agents and applies safe, verifiable self-improvement patches.

Core Features & Use Cases

  • Self-diagnosis of performance issues against defined metrics.
  • Patch generation and safe application with safety guards (circuit breaker, blast radius, test gate).
  • Evolution logging for traceability and rollback capability.
  • Minimal PoC implementations that demonstrate the loop from diagnosis to verification.

Quick Start

Run a diagnosis on the agent’s current metrics, generate a safe patch, apply it, verify with tests, and review the Evolution Log.

Frequently Asked Questions about agentforge-evolution

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

FAQPage Schema
How do I safely apply self-improvement patches to AI agents?

To safely apply self-improvement patches to AI agents, diagnose faults against defined metrics, generate a patch, apply it with safety guards like a circuit breaker, verify with test gates, and review the evolution log.

How does a circuit breaker mechanism work for AI agent self-evolution?

A circuit breaker mechanism for AI agent self-evolution halts patch application automatically if blast-radius checks or test gates detect critical faults, ensuring changes remain traceable and reversible through evolution logging.

What is the best way to trace and rollback AI agent evolution changes?

The best way to trace and rollback AI agent evolution changes is by maintaining an Evolution Log that records every patch application, enabling verifiable reversals when safety gates or test verifications fail.

Can I use automated patch management for AI agents in Platform-Harness environments?

Yes, you can use automated patch management for AI agents in Platform-Harness environments across the full lifecycle from specification to deployment, utilizing blast-radius checks and evolution logs for safe changes.

What are the limitations of relying on self-diagnosis for AI agent patch generation?

Relying on self-diagnosis for AI agent patch generation requires defined performance metrics to detect faults accurately; without strict test gates and blast-radius checks, generated patches risk unverified runtime failures.