Evolutionary Model
v1.0.0Framework for building AI agents that evolve with their owner. Use when: setting up a new agent from scratch, onboarding a team to AI-native workflow, explai...
⭐ 1· 44·0 current·0 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
The name/description (a framework for building evolving agents) matches the SKILL.md: it is documentation and policy for organizing files, memories, skills, and protocols. No binaries, installs, or credentials are requested — appropriate for a purely instructional framework.
Instruction Scope
The runtime instructions explicitly instruct creating, reading, and writing files (SOUL.md, USER.md, memory/, skills/, ~/clawd/skills/, etc.). That behavior is central to the stated purpose (persistent agent memory and skills). However, the skill metadata declares no required config paths while the instructions assume specific filesystem locations; this mismatch is worth noting because users should expect the agent to access and mutate files in the user's home or project directory.
Install Mechanism
No install spec and no code files — lowest-risk delivery model. Nothing is downloaded or written by an installer; the skill is a set of instructions only.
Credentials
No environment variables or credentials are requested. The SKILL.md recommends using config.json or env vars for individual skills, which is reasonable guidance and does not itself require credentials.
Persistence & Privilege
The skill advocates persistent state stored as files and a skills directory; it does not request 'always: true' or other elevated platform privileges. Users should recognize this will cause the agent to create and maintain files with long-lived context on disk, which has security and privacy implications even though the skill itself does not ask for extra platform privileges.
Assessment
This is a documentation-only skill describing a file-backed, persistent agent architecture — it does not install code or request secrets. Before using: be aware the agent will read and write persistent files (~/clawd, memory/, skills/) that can contain sensitive context; review and control where those files are stored and who can access them. Do not store tokens/passwords in plain text files; prefer config.json with appropriate filesystem permissions or a secrets manager. If you let the agent run scripts from skills/scripts, inspect those scripts first for network calls or dangerous commands. Finally, confirm the agent runtime’s filesystem and autonomy policies (what it’s allowed to read/write/execute) before enabling this behavior.Like a lobster shell, security has layers — review code before you run it.
frameworkvk970pfrkctexmt4n52qt86pve983z456latestvk970pfrkctexmt4n52qt86pve983z456personal-osvk970pfrkctexmt4n52qt86pve983z456
License
MIT-0
Free to use, modify, and redistribute. No attribution required.