Evolve AI Skills with Lamarck: Dynamic Agent Skill Enhancement

Evolve AI Skills with Lamarck: Dynamic Agent Skill Enhancement

The Evolve AI Skills with Lamarck provides a mechanism to dynamically enhance the skills of AI agents, such as Claude Code, Cursor, and Codex, through a continuous feedback and evolution process. This capability leverages Lamarckian evolution to intelligently adjust and improve AI skills based on real-world usage data.

What this skill does

• Monitors real-time skill invocations and records their performance, utilizing a PostToolUse hook to log each invocation with the specific skill's genome hash for version tracking.
• Executes a light evaluation loop after each invocation that doesn’t reload the SKILL.md file every turn, which assesses the skill using dynamic rubrics defined per skill.
• Manages skill evolution through a trust ladder configuration to determine if skills can be auto-edited, evolved, suggested for change, or merely observed.
• Promotes the evolution of skills by writing back learned experiences into their "genomes", adapting the criteria for evaluation and removing unused features over time.

Who it is for

Developers and teams who utilize AI coding agents like Claude Code, Cursor, and Codex will benefit from this skill, as it automates the process of evaluating and evolving AI-generated skills based on actual application and performance data.

Use cases

• Review and optimize AI skill performance after observing a significant amount of evidence suggesting potential improvements.
• Audit the skill ledger to maintain an updated record of skill performance and evolution history.
• Manage the evolution whitelist to determine which skills are eligible for automatic modification based on performance data.
• Switch Lamarck trigger modes to adapt the evolution process as per organizational needs.

Technical details

• Integrates with AI agents like Claude Code, Cursor, and Codex for skill monitoring and evolution.
• Implements a PostToolUse hook to log invocation data in `data/pending.jsonl` and a Stop hook for evaluation recorded in `data/ledger.jsonl`.
• Utilizes dynamic rubrics in a git-versioned format to define performance criteria per skill.
• Configurations such as the trust ladder are managed through the `config.json` file.

Source & Licence

This package is built on open-source work published by newdee (newdee/lamarck-skill) and distributed under MIT. The original licence text and copyright notice are included in your download.

Personal and commercial use, modification and redistribution are permitted, provided the original copyright and licence notice are retained.

Your purchase covers curation, licence verification, packaging, documentation and instant delivery. It does not grant exclusive rights to the underlying open-source code, which remains available under its original licence.

Delivery & Support

• Delivery: instant — a secure download link is emailed to you as soon as payment is confirmed.
• Format: ZIP archive containing the skill files, documentation and the original licence.
• Support: support@mcpcart.com — we aim to reply within 2 business days.
• Updates: updates are included only where stated on this page.

Refunds

This is a digital product delivered immediately after purchase. By completing your order you request immediate delivery and acknowledge that, once the download has been accessed, the statutory right to cancel no longer applies to the extent permitted by law. Refund requests are handled in accordance with our published Refund Policy.

Claude, Codex, Gemini and Cursor are trademarks of their respective owners. MCP Cart is an independent marketplace and is not affiliated with, endorsed by, or sponsored by any of them. Compatibility references describe interoperability only.