ensemble-content-scorer
Pass
Audited by Gen Agent Trust Hub on Nov 6, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because user-provided input is interpolated directly into the system prompts used to query external LLMs.
- Ingestion points: User-provided
ideaandideasparameters inscripts/score_content.py. - Boundary markers: No delimiters or specific instructions to ignore embedded commands are used to isolate the untrusted input from the rest of the prompt template.
- Capability inventory: The skill performs network operations (API calls to Anthropic, OpenAI, and Google) and has the ability to write reports to the local file system when the
--outputflag is used. - Sanitization: There is no evidence of input sanitization or validation of the content ideas before they are processed by the script.
- [EXTERNAL_DOWNLOADS]: The skill lists several external dependencies required for interacting with AI services.
- Fetches official packages including
anthropic,openai, andgoogle-generativeaifrom established registries. These are official libraries provided by well-known technology organizations.
Audit Metadata