investigate-create-doc

Pass

Audited by Gen Agent Trust Hub on Mar 3, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection during the exploration of the codebase. \n
  • Ingestion points: Codebase content is read and processed in Phase 4 ('Investigate Each Area') by subagents. \n
  • Boundary markers: The subagent prompt template interpolates investigation questions directly with codebase context without using specific boundary markers or 'ignore embedded instructions' warnings. \n
  • Capability inventory: The skill uses the 'Task' tool to spawn subagents which have the ability to read the filesystem and write markdown documentation to the 'docs/' directory. \n
  • Sanitization: There is no evidence of sanitization, escaping, or validation of the data read from the codebase before it is used by subagents or included in documentation. \n- [COMMAND_EXECUTION]: To fulfill the investigation requirements (tracing dependencies, finding consumers), the skill relies on the agent's ability to execute search commands and file system operations.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 3, 2026, 03:29 AM