brief-ku
Pass
Audited by Gen Agent Trust Hub on Sep 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes free-form application ideas from users, which is an inherent ingestion surface for instructions embedded in data.
- Ingestion points: The agent accepts and analyzes application descriptions provided by the user in SKILL.md.
- Boundary markers: There are no explicit instructions or delimiters defined to isolate user-provided text from the system instructions.
- Capability inventory: The skill uses interactive clarification tools such as AskUserQuestion, request_user_input, and question to interact with the user.
- Sanitization: No specific sanitization or validation logic is prescribed for the input text before processing.
- [EXTERNAL_DOWNLOADS]: The skill provides installation instructions that fetch resources from an external GitHub repository.
- The README.md specifies an installation command that downloads the skill from the author's repository at github.com/Adytm404/skills.
Audit Metadata