recall

Pass

Audited by Gen Agent Trust Hub on Jun 13, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No malicious patterns detected. The skill implements a logical workflow for query expansion, searching via a dedicated tool, and reranking results based on user intent.
  • [INDIRECT_PROMPT_INJECTION]: The skill retrieves and processes memories which may contain untrusted data.
  • Ingestion points: The output of the mcp__plugin_origin_origin__recall tool is processed for reranking and display (SKILL.md).
  • Boundary markers: No specific delimiters or safety instructions are defined for the rendered memory content.
  • Capability inventory: The skill is restricted to memory search functions via mcp__plugin_origin_origin__recall as specified in allowed-tools.
  • Sanitization: No explicit sanitization or filtering is performed on retrieved memory content before processing or display.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 13, 2026, 07:51 AM
Security Audit — agent-trust-hub — recall