dojo-andy-grove

Pass

Audited by Gen Agent Trust Hub on Sep 22, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user input (managerial questions and business situations) to route the request to a specific expert persona and generate a response based on topic frameworks.\n
  • Ingestion points: User input provided during interaction is used for classification and routing as described in SKILL.md.\n
  • Boundary markers: The skill includes grounding instructions to 'Answer in that expert's voice using only the substance in the files you loaded' and 'Never invoke a framework that isn't in the topic files,' which help prevent instructions in user data from overriding the agent's core behavior.\n
  • Capability inventory: The skill does not define or use any tools, scripts, or external capabilities; it is a purely informational skill.\n
  • Sanitization: No explicit input sanitization or validation is implemented for user-provided situational descriptions.\n- [NO_CODE]: The skill is composed exclusively of Markdown documentation and persona instructions. It does not include any executable scripts, binaries, or platform-specific tool definitions.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 22, 2026, 03:31 PM
Security Audit — agent-trust-hub — dojo-andy-grove