Agentic Loop Memory Server
Warn
Audited by Socket on Mar 15, 2026
1 alert found:
SecuritySecuritySKILL.md
MEDIUMSecurityMEDIUM
SKILL.md
SUSPICIOUS: the skill's memory-server purpose is plausible, but its actual footprint is broader than a simple note store. It externalizes extensive task history, feeds that state back into agent actions, promotes autonomous looping, and appears to expose 'System Instructions' to an unverifiable third-party MCP server with weak provenance.
Confidence: 82%Severity: 72%
Audit Metadata