honcho

Pass

Audited by Gen Agent Trust Hub on Sep 8, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill manages a memory system that ingests user observations and conclusions, which are later injected into the agent's system prompt. While this creates a surface for indirect prompt injection, the documentation explicitly notes sanitization measures such as stripping XML/HTML tags and escaping delimiter sequences to mitigate this risk.
  • [EXTERNAL_DOWNLOADS]: The skill requires the installation of the 'honcho-ai' Python package, which is the official client for the described memory service.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 8, 2026, 06:56 AM
Security Audit — agent-trust-hub — honcho