creativeclaw-find-examples
Pass
Audited by Gen Agent Trust Hub on Sep 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests external, natural-language prompts which can potentially be used to influence agent behavior.
- Ingestion points: The
get_exampletool defined inSKILL.mdretrieves agent-ready prompts and generation hints from an external catalog. - Boundary markers: While
SKILL.mdprovides instructions to adapt the examples to the user's specific subject, it does not mandate the use of explicit boundary markers or delimiters for the external content. - Capability inventory: The skill environment includes a range of tools for media assembly, job management, and file uploads (e.g.,
merge_audios,check_job,upload_asset) as detailed in the reference files. - Sanitization: There is no mention of sanitization or validation of the retrieved prompt content before it is processed by the agent.
Audit Metadata