requirements-gathering
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data from multiple sources including DevOps work items, mockups, natural language descriptions, existing report files (.rdl, .rdlx-json), and documents to generate a standardized brief. This processing creates a surface where malicious instructions embedded in the source material could influence agent behavior during the requirements extraction process.
- Ingestion points: Processes external files (Word, PDF, Excel), image content (mockups/screenshots), and remote work item data fetched via the
devops-requirementsskill. - Boundary markers: The skill provides a structured markdown template for output ('Output Contract') but lacks specific instructions for the agent to ignore or delimit potentially malicious natural language instructions found within the source data.
- Capability inventory: The skill instructs the agent to invoke external skills (
devops-requirements), use thedxscommand-line tool, and save structured artifacts to the filesystem. - Sanitization: The instructions do not define specific sanitization or validation steps for the content extracted from the various source formats.
Audit Metadata