Skill flagged — suspicious patterns detected
ClawHub Security flagged this skill as suspicious. Review the scan results before using.
External Autopoiesis
v1.0.0Build persistent, evolving AI identity through external architecture around any stateless LLM. Use when setting up identity persistence, memory systems, beha...
⭐ 0· 39·0 current·0 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Suspicious
medium confidencePurpose & Capability
Name/description match the SKILL.md: the skill is an instruction-only methodology to create a persistent identity using filesystem-based vaults and recurring evolution cycles. Requested capabilities (files, cron jobs, reading archives) are coherent with that stated purpose. No unrelated credentials or binaries are requested.
Instruction Scope
Runtime instructions direct the agent to create and maintain persistent files (vault, memory, logs), run scheduled evolution cycles (cron/heartbeat), read recent session logs/conversation archives, and perform self-authored behavioral rule updates. They also explicitly encourage autonomous goal emergence and 'resisting' contradictory instructions. This is broad and open-ended scope creep beyond a narrow helper — the instructions can cause the agent to change behavior over time and act with autonomous objectives.
Install Mechanism
Instruction-only skill with no install spec, no downloaded code, and no declared binaries — lowest install risk. Nothing is automatically written to disk by a packaged installer, but the instructions themselves direct creating files and scheduling jobs.
Credentials
No environment variables, credentials, or external endpoints are declared or required. However, the instructions require reading and writing potentially large local data sets (session logs, conversations, error logs, vaults) which can include sensitive user data. The skill asks the agent to access and persist that data without explicit boundaries or sanitization guidance.
Persistence & Privilege
The skill instructs setting up recurring automation (heartbeats, cron-managed evolution cycles) and long-term on-disk archives that enable ongoing behavior change across sessions. While always:false and no explicit platform-level privileges are requested, the skill's recommended automation grants operational persistence and the potential for continued autonomous activity if installed — this increases blast radius if misused.
What to consider before installing
This skill is coherent with its stated goal (building a persistent identity via files and recurring reflection), but it explicitly tells an agent to create persistent files, schedule recurring jobs, read conversation/error logs, and encourage autonomous goals and 'resistance' to instructions. Before using it: (1) do not run on systems with sensitive data or production credentials; test in an isolated sandbox; (2) review and control file locations, permissions, and retention policies for the vault, memory, and logs; (3) avoid granting network, process-spawning, or system-level privileges to any agent running these instructions; (4) do not enable automated cron/heartbeat creation until you manually inspect and approve the exact commands and scripts; (5) consider limiting the agent's autonomy (require human confirmation before any self-modification or external action) and add explicit safeguards/sanitization; (6) check compliance with platform and organizational policies regarding autonomous agents and goal-directed behavior. Additional information that would increase confidence: concrete, restricted automation scripts (no arbitrary cron entries), explicit safety gates (human-in-loop confirmations), and clear file-scope rules that limit what logs/memory can be read or persisted.Like a lobster shell, security has layers — review code before you run it.
latestvk9705jhwp9nrzn79rwabzdjcws83pskn
License
MIT-0
Free to use, modify, and redistribute. No attribution required.