interview-conducting
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- Indirect Prompt Injection (LOW): The skill is designed to ingest and process untrusted data from external stakeholders via the
AskUserQuestiontool. This creates a potential vulnerability where an attacker posing as a stakeholder could provide malicious instructions embedded in their responses. - Ingestion points: Stakeholder responses captured through the
AskUserQuestiontool and processed in the 'Requirement Extraction' and 'Interview Summary' phases. - Boundary markers: The skill lacks explicit delimiters or instructions to the LLM to ignore or sanitize commands embedded within stakeholder input.
- Capability inventory: The skill possesses the
Writetool to save data to the local filesystem and theTasktool to spawn sub-agents, which could be misused if the agent is manipulated by injected instructions. - Sanitization: No sanitization or validation logic is present to filter stakeholder input before it is used to generate YAML-formatted requirements or summary files.
Audit Metadata