iterate-research-questions
Pass
Audited by Gen Agent Trust Hub on Oct 19, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill reads from research documents and ticket files which may contain untrusted data that could influence the agent's behavior.
- Ingestion points: Content is read from a
docPathand an optional ticket file in steps 1 and 2. - Boundary markers: The instructions lack specific delimiters or warnings to ignore instructions embedded in the input data.
- Capability inventory: The skill allows the agent to read and write files and perform directory listings within the task environment.
- Sanitization: There is no requirement to sanitize or validate input content before it is processed.
- [COMMAND_EXECUTION]: The skill explicitly instructs the agent to execute shell commands to navigate the filesystem.
- Evidence: The step for locating documents specifies running
ls -La .humanlayer/tasks/<task slug>to identify artifacts.
Audit Metadata