gam-agentic-memory

Pass

Audited by Gen Agent Trust Hub on May 18, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [PROMPT_INJECTION]: The skill processes untrusted external data (PDFs, JSONL, videos) via the wf.add() command to build hierarchical memory. This creates a surface for indirect prompt injection where instructions hidden in the files could influence agent behavior. Ingestion points: Content added via wf.add() in SKILL.md. Boundary markers: Absent. Capability inventory: Querying memory via wf.request() or SDK. Sanitization: Absent.
  • [EXTERNAL_DOWNLOADS]: The documentation provides standard pip install commands for local development and dependency management.
  • [SAFE]: The skill demonstrates secure credential handling by instructing users to use environment variables like GAM_API_KEY instead of hardcoding secrets.
Audit Metadata
Risk Level
SAFE
Analyzed
May 18, 2026, 01:48 AM
Security Audit — agent-trust-hub — gam-agentic-memory