honcho

Pass

Audited by Gen Agent Trust Hub on Apr 19, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill requires the honcho-ai Python package, which is the official client library for the Honcho memory service.
  • [SAFE]: Data transmission to app.honcho.dev is the core intended functionality of the skill, used for storing and retrieving user observations and conclusions for the Hermes agent.
  • [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface due to its memory retrieval capabilities.
  • Ingestion points: Data enters the context via the honcho_search and honcho_context tools (SKILL.md).
  • Boundary markers: The documentation does not specify the use of delimiters or specific instructions to the model to ignore instructions embedded in the retrieved memory.
  • Capability inventory: The skill provides tools for semantic search, dialectic reasoning, and writing conclusions to a persistent store.
  • Sanitization: There is no evidence of sanitization or filtering of the content retrieved from the Honcho API before it is presented to the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 19, 2026, 07:17 AM