Enterprise Memory
Security checks across static analysis, malware telemetry, and agentic risk
Overview
The skill implements a local, file-based Enterprise Agent Memory (EAM) system and its bundled Python scripts match the described purpose; nothing requested is disproportionate, but note that some scripts write/rename local project files and an optional compression step can call an external LLM API if configured.
This skill appears to do what it claims: it manages EAM projects under ~/.openclaw/EAM-projects using local Python scripts. Before installing/using: (1) back up any existing ~/.openclaw/EAM-projects because scripts (e.g., renumber_projects) rename directories and write files; (2) review and run scripts in dry-run mode where available; (3) avoid providing arbitrary external LLM API endpoints to the compress step unless you trust the endpoint, since that may transmit project logs/state/decisions; (4) note high-risk actions (setting DONE/ARCHIVED, changing owner) require --confirm and are guarded in code. No credentials or remote installs are requested by the skill itself.
SkillSpector
SkillSpector findings are pending for this release.
Static analysis
No static analysis findings were reported for this release.
VirusTotal
No VirusTotal findings