Self-Evolve Skill

Agent自进化机制。三层架构（记忆→技能→规范），三种触发（关键词+踩坑+复盘），用户审批+Darwin评分双重质控。用得越多越懂你。触发词：自进化、进化、evolve、记住、以后、永远、下次、踩坑经验。

Before installing: 1) Understand this skill will persist rules into workspace files and run git commits (MEMORY.md, SKILL.md, CLAUDE.md, evolution logs). Ensure you want those persistent changes and have backups. 2) Confirm that git and Python are available in your agent environment and that you accept the skill reading other skills' files (it references darwin-skill). 3) Be cautious about allowing L3/global changes — although SKILL.md requires double confirmation, approved L3 writes can change behavior for all bots. 4) If you want lower risk, disable autonomous invocation or require manual invocation/explicit confirmations for any writes to CLAUDE.md or other skills. 5) Test in a sandbox workspace first and review commits created by the skill to confirm it only writes expected content.

Type: OpenClaw Skill Name: shike-self-evolve Version: 2.1.0 The skill bundle implements a 'self-evolution' mechanism that allows the agent to modify its own instruction files (SKILL.md, MEMORY.md, and the global CLAUDE.md) based on user input and execution history. While the instructions in SKILL-full.md include explicit security boundaries and require user approval for changes, the capability for self-modification of core system prompts (L3 evolution) represents a high-risk attack surface for persistent prompt injection and behavioral manipulation. The Python logic in scripts/evolution_engine.py appears benign, focusing on keyword detection and failure metrics without evidence of data exfiltration or unauthorized network access.