onboard
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONCREDENTIALS_UNSAFE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill collects user responses to ten discovery questions and interpolates them directly into generated configuration files (such as SOUL.md and AGENTS.md). This creates an indirect prompt injection surface where user-supplied content becomes part of the agent's persistent identity and behavioral instructions.\n
- Ingestion points: User answers to the identity, workflow, and tool-related discovery questions defined in SKILL.md.\n
- Boundary markers: The skill does not provide instructions to the agent to use delimiters or ignore instructions within the user's answers when writing to configuration files.\n
- Capability inventory: The skill utilizes file-system write access to generate and store configuration files across multiple agent environments.\n
- Sanitization: No explicit validation or filtering logic is provided to the agent for processing user input before it is written to the configuration files.\n- [COMMAND_EXECUTION]: The skill includes automated testing of tool integrations, which involves executing verification commands. These actions are performed with user transparency and explicitly require explaining security implications in plain language before execution.\n- [CREDENTIALS_UNSAFE]: The skill follows security best practices for credential management by explicitly directing the agent to store API keys in local environment files (.env.local) and providing users with step-by-step instructions to prevent hardcoding or accidental exposure of secrets.
Audit Metadata