Skill flagged — suspicious patterns detected
ClawHub Security flagged this skill as suspicious. Review the scan results before using.
four-layer-memory
v1.0.1四层记忆架构。身份层、工作记忆层、短期日志层、长期存储层分层管理，让记忆既可持续累积，又不会把上下文一次性塞满。
⭐ 0· 73·0 current·0 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Suspicious
high confidencePurpose & Capability
Name/description (four-layer memory architecture) align with the content of SKILL.md: it documents identity, working, short-term, and long-term layers and how to use them.
Instruction Scope
SKILL.md recommends running python3 scripts/personal_ai_memory.py with commands and references directories (identity/, working-memory/, short-term-logs/, archive/) — but no code files or config paths are included in the package and the skill declared no required paths or env vars. That mismatch means the instructions either expect external files to exist or implicitly require the agent to access/modify local filesystem state; both are not declared.
Install Mechanism
No install spec and no code files — lowest risk for automatic code injection. However, the recommended runtime commands would execute a local Python script if present; the skill itself does not install any binaries.
Credentials
No environment variables, credentials, or config paths are requested. That is proportionate to an instruction-only documentation skill. The concern is the hidden requirement for local scripts/directories which were not declared.
Persistence & Privilege
always is false and the skill is user-invocable; it does not request persistent privileged presence or modification of other skills' configs.
What to consider before installing
This skill is mostly documentation of a memory-layer design, but it instructs the agent to run local Python scripts and read/write named directories that are not included or declared. Before installing or enabling it: 1) ask the publisher for the referenced scripts (scripts/personal_ai_memory.py) and inspect them — do not run them blind; 2) confirm which filesystem paths the skill will use and that those paths don't contain sensitive data; 3) if you want to test, run any provided scripts in a sandboxed environment or container; 4) disable autonomous invocation or limit the agent's shell/file permissions until you verify the code; and 5) if the author cannot supply the scripts or a trustworthy source, treat this skill as incomplete and avoid enabling it.Like a lobster shell, security has layers — review code before you run it.
latestvk97a9t714xbyh367cvxnk7dbdx83y3nj
License
MIT-0
Free to use, modify, and redistribute. No attribution required.