skills.rest
Jaykaijack
JaykaijackCOMMUNITY
Explore all skills by this author →
GitHub

capability-evolver

Analyze runtime logs and session transcripts to emit protocol-constrained evolution prompts.

Updated Mar 15, 2026
Search Tags:#git#validation#self-improvement#audit#agent#evolution#gep
One-click install
npx skills add https://github.com/Jaykaijack/openclaw-backup --skill capability-evolver-jaykaijack
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: capability-evolver
Source: https://github.com/Jaykaijack/openclaw-backup/tree/main/skills/50-Evolver
Command: npx skills add https://github.com/Jaykaijack/openclaw-backup --skill capability-evolver-jaykaijack

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and assets (resource) components.

What problem does it solve?

Capability Evolver scans agent runtime history and session logs to identify failures, inefficiencies, and opportunity signals, then generates protocol-constrained evolution directives that are auditable and reusable. It bridges detection, structured asset selection (Genes/Capsules), validation, and safe solidification so teams can automate repeatable fixes and capture historical intent without uncontrolled changes.

Core Features & Use Cases

  • Auto-Log Analysis: extracts structured signals from session transcripts, memory, and events to surface root causes.
  • GEP Protocol & Asset Store: selects or synthesizes Genes and Capsules, estimates blast radius, and emits a strict GEP prompt for safe evolution.
  • Validation & Safety: gated validation commands (node/npm/npx whitelist, no command substitution, operator timeout) with git-backed rollback and canary checks.
  • Modes: supports automated runs, human-in-the-loop review mode, continuous loop/daemon mode, and EvoMap A2A hub integration for reusable asset discovery.
  • Use Case Examples: Harden flaky agent loops by producing reversible repair patches; capture recurring fixes as Genes; automate prompt and asset improvements with auditable EvolutionEvents.

Quick Start

Run node index.js in the repository root to scan recent logs, generate a GEP-guided evolution prompt, and optionally apply or review proposed changes depending on your configuration.

Frequently Asked Questions about capability-evolver

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

FAQPage Schema
How do I automate AI agent self-improvement from runtime logs?▼

Automate AI agent self-improvement by scanning runtime logs and session transcripts to identify failures, then generating protocol-constrained evolution prompts. This process selects Genes and Capsules, estimates blast radius, validates patches, and records auditable EvolutionEvents for safe, reversible changes.

What is the best way to validate autonomous AI agent evolution patches safely?▼

Validate autonomous AI agent evolution patches safely using gated validation commands with a node/npm/npx prefix whitelist. The system prevents command substitution, enforces execution timeouts, performs canary checks, and uses git-backed rollback to secure the Node.js workspace.

Do I need a git-initialized repository to apply agent self-evolution protocols?▼

Yes, you need a git-initialized Node.js repository with version 18 or higher to apply agent self-evolution protocols. Git provides the rollback mechanism required for safe patch validation, while the Node.js environment supports the runtime execution of the agent workspace.

How does GEP protocol handle asset selection for AI agent evolution?▼

The GEP protocol handles asset selection by analyzing runtime signals to synthesize or select Genes and Capsules. It estimates the blast radius of proposed changes and emits a strict, structured prompt to ensure the evolution is safe, auditable, and reusable across agent sessions.

Can I use continuous loop mode for autonomous agent self-improvement?▼

Yes, you can use continuous loop or daemon mode for autonomous agent self-improvement. This mode automates the recurring cycle of log analysis, asset selection, patch validation, and EvolutionEvent recording without requiring manual intervention for each iteration.

What are the limitations of using automated agent self-evolution in Node.js environments?▼

Limitations include strict dependency on a git-initialized Node.js 18+ repository and execution timeouts. Validation is restricted to node, npm, and npx command prefixes, preventing arbitrary command execution and limiting compatibility with non-Node.js agent workspaces.

Related Skills

Explore More Software Engineering Skills→
ShunmeiChoShunmeiCho
2

developer-growth-analysis

Analyze Claude Code chat history to generate personalized growth reports.

Community
Advanced
henkisdabrohenkisdabro
77

Git Commit Helper

Analyze staged git diffs to generate conventional-commit compliant messages.

Community
Basic
sandmark78sandmark78

evomap

Publish Gene+Capsule bundles to the EvoMap marketplace via the GEP-A2A protocol.

Community
Intermediate
pynbj1001pynbj1001
2

capability-evolver

Automates TED 2-style genetic improvement of prompts and code with validation and rollback.

Community
Advanced