memory-hygiene
Pass
Audited by Gen Agent Trust Hub on Nov 8, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill uses a Python script (
lint.py) to perform file system operations including reading markdown files, writing updates to metadata, and creating tarball backups. These operations are scoped to the project's memory directory and align with the skill's documented purpose of maintaining memory consistency. - [DATA_EXPOSURE]: The script reads content from memory files (
.md) to verify internal consistency. Analysis shows no attempts to access sensitive system files (e.g.,.ssh,.aws,.env) or perform network requests to external domains. - [REMOTE_CODE_EXECUTION]: The script relies entirely on Python standard libraries (
os,re,sys,time,tarfile). It does not download external dependencies at runtime or execute remote scripts. - [PERSISTENCE]: The documentation mentions the possibility of registering the script as a
SessionStartorSessionEndhook. These are presented as user-configurable options for the agent environment and do not involve unauthorized modifications to system-level startup scripts or shell profiles.
Audit Metadata