prompt-lookup
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes prompt templates retrieved from the external prompts.chat service.
- Ingestion points: External prompt data is ingested via search_prompts and get_prompt tools described in SKILL.md.
- Boundary markers: Absent; the instructions do not include markers to distinguish external content from the system prompt.
- Capability inventory: The capabilities defined in SKILL.md (search_prompts, get_prompt, improve_prompt) are restricted to text handling and do not include high-risk operations like file system access or shell command execution.
- Sanitization: Absent; there is no specified validation or filtering for the retrieved prompts before they are used or enhanced.
Audit Metadata