ask-clarifying-questions
Pass
Audited by Gen Agent Trust Hub on Oct 12, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to ingest and act upon data from potentially untrusted external sources, creating a surface for indirect prompt injection.
- Ingestion points: The process explicitly directs the agent to read project-specific files such as
AGENTS.mdandCLAUDE.md, as well as general project files, git status, diffs, and user-provided answers to interview questions. - Boundary markers: The skill lacks explicit instructions for the agent to distinguish between legitimate data and malicious instructions that might be embedded within the project files or user responses.
- Capability inventory: The skill utilizes file read capabilities and interactive communication tools including
AskUserQuestionandrequest_user_inputto influence the agent's subsequent coding and design decisions. - Sanitization: No sanitization, validation, or filtering mechanisms are defined for the content retrieved from these external files or user interactions before the agent uses them to 'govern implementation'.
Audit Metadata