skills.rest
ste-bah
ste-bahCOMMUNITY
Explore all skills by this author →
GitHub

evolve-agent

Analyzes agent traces and applies FIX, DERIVED, CAPTURED evolution suggestions.

Updated Aug 27, 2026
Search Tags:#automation#versioning#analysis#agent#evolve#memorygraph
One-click install
npx skills add https://github.com/ste-bah/archon --skill evolve-agent
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: evolve-agent
Source: https://github.com/ste-bah/archon/tree/main/.claude/agents/.claude/skills/evolve-agent
Command: npx skills add https://github.com/ste-bah/archon --skill evolve-agent

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

The Evolution workflow helps teams improve custom agents by surfacing past analyses, consolidating evolution suggestions, and guiding precise code changes.

Core Features & Use Cases

  • Surface and deduplicate evolution suggestions from agent analyses and traces.
  • Present FIX, DERIVED, and CAPTURED evolution paths with context and impact assessments.
  • Apply approved changes to agent definitions, generate version history, and record lineage in MemoryGraph.

Quick Start

Provide an agent name and trigger the evolution workflow to review available improvements for that agent.

Frequently Asked Questions about evolve-agent

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

FAQPage Schema
How do I automate agent evolution using past traces?▼

Agent evolution automates improvements by analyzing recent traces and MemoryGraph analyses to surface deduplicated suggestions. It identifies actionable changes and guides precise code modifications for custom agents.

What are FIX, DERIVED, and CAPTURED evolution paths?▼

FIX, DERIVED, and CAPTURED are evolution pathways that categorize agent improvement suggestions with context and impact assessments. They guide how changes are applied to agent definitions and recorded in version history.

How do I deduplicate agent evolution suggestions from analyses?▼

Deduplicating agent evolution suggestions involves consolidating past analyses and traces to present unique improvements. This workflow surfaces ranked suggestions, preventing redundant code changes and ensuring precise agent updates.

Does evolving a custom agent require MemoryGraph?▼

Evolving a custom agent requires access to MemoryGraph analyses, traces, and agent definitions. MemoryGraph is necessary to execute or simulate evolutions and record the lineage of applied changes.

Can I review context before applying agent evolution changes?▼

You can review context before applying changes because the workflow presents deduplicated evolution suggestions with impact assessments. This ensures approved changes are safely applied to agent definitions with generated version history.

Related Skills

Explore More Software Engineering Skills→
robertpellonirobertpelloni
2

tracing-knowledge-lineages

Trace idea lineage through decision records and version history.

Community
Intermediate
struktured-labsstruktured-labs

restore

Restore LMMS projects to a previous version using a version hash.

Official
Advanced
GuitarAlchemistGuitarAlchemist

self-improve-loop

Analyze execution traces and promote patterns across GA, ix, and TARS systems.

Official
Advanced
GDSDNGDSDN
3

dev-suggest-refactoring

Analyze codebases to suggest automated refactoring opportunities using AST processing and code metrics.

Community
Advanced