brain-taxonomist

Pass

Audited by Gen Agent Trust Hub on Aug 29, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill determines filing paths by analyzing user-provided content such as notes and brain pages. This creates an attack surface where instructions hidden within the processed documents could attempt to manipulate the agent's decision-making process. The skill mitigates this by involving the user in ambiguous or low-confidence decisions.
  • Ingestion points: The skill ingests data from new brain pages, bulk imports, and notes as defined in the 'Decision Protocol' section of SKILL.md.
  • Boundary markers: There are no explicit delimiters or boundary markers mentioned to isolate untrusted user data.
  • Capability inventory: The skill calls local CLI tools (gbrain and jq) for schema lookups and consistency checks, but lacks network or write permissions.
  • Sanitization: No sanitization or validation of the input text is described in the prompt logic.
  • [COMMAND_EXECUTION]: The skill instructions include executing local command-line tools like 'gbrain' and 'jq' to interact with the active schema pack. These tools are used for retrieving configuration and performing diagnostic checks within the environment.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 29, 2026, 06:18 PM
Security Audit — agent-trust-hub — brain-taxonomist