vector-search

Pass

Audited by Gen Agent Trust Hub on Jun 22, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides documentation for performing vector and hybrid searches using a local memory daemon. No malicious code, obfuscation, or unauthorized access patterns were identified.
  • [COMMAND_EXECUTION]: The instructions guide the agent to execute memory-daemon commands for memory retrieval. These commands are local to the system and serve the skill's primary function of semantic search.
  • [DATA_EXFILTRATION]: Network activity is restricted to a gRPC connection on the local loopback address http://[::1]:50051. This communication is intended for internal data retrieval from the memory daemon and does not involve sending data to external or untrusted destinations.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 22, 2026, 02:44 AM
Security Audit — agent-trust-hub — vector-search