brainstorming
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted data from the local repository environment which could harbor malicious instructions. * Ingestion points: The workflow requires the agent to read local files, documentation, and recent commit messages to gain context (SKILL.md). * Boundary markers: There are no explicit delimiters or warnings provided to the agent to ignore instructions found within the ingested project data. * Capability inventory: The skill allows the agent to write design documents to the system, execute git commits, and trigger the writing-plans skill. * Sanitization: The workflow lacks specific validation or sanitization steps for the content extracted from external project files.
- [COMMAND_EXECUTION]: The skill involves the execution of shell-level repository management commands. * Evidence: The documentation in SKILL.md explicitly directs the agent to commit design documents to the git repository using local tooling.
Audit Metadata