jobs-to-be-done
Pass
Audited by Gen Agent Trust Hub on Sep 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [SAFE]: The skill contains no executable scripts, binaries, or obfuscated content. It functions as a prompt-based orchestration framework using established business analysis methodologies (Clayton Christensen's JTBD).
- [INDIRECT_PROMPT_INJECTION]: The skill instructions define an ingestion surface for potentially untrusted data to ground the analysis in real evidence.
- Ingestion points: Stage 2 subagents are instructed in
SKILL.mdto use retrieval tools to gather data from external sources such as web forums, reviews, and support logs. - Boundary markers: The instructions explicitly require subagents to return 'structured data, not prose for a human,' which acts as a structural boundary to prevent the execution of embedded instructions.
- Capability inventory: The skill possesses the capability to write the final synthesized report to a local markdown file.
- Sanitization: No specific sanitization or escaping logic is defined for the retrieved text beyond the requirement for structured data output.
- [COMMAND_EXECUTION]: While the skill mentions using 'retrieval tools,' it does not provide specific shell commands or scripts, relying instead on the agent's built-in tool environment for information gathering.
Audit Metadata