blog-cannibalization
Pass
Audited by Gen Agent Trust Hub on Jun 13, 2026
Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes untrusted data from local blog files (.md, .mdx, .html). Maliciously crafted content within these files could attempt to subvert the keyword extraction or semantic clustering logic.
- Ingestion points: Reads content from files in a user-specified directory (SKILL.md).
- Boundary markers: None identified to separate file content from the agent's instructions.
- Capability inventory: The skill uses WebFetch to communicate with an external API and generates reports based on processed data.
- Sanitization: No sanitization or validation of the extracted keyword strings is performed before they are used in semantic analysis or API requests.
- [EXTERNAL_DOWNLOADS]: The skill performs network operations to the DataForSEO API (api.dataforseo.com) when the API mode is enabled. This is used to fetch SEO metrics like search volume and keyword rankings from an established SEO service provider.
Audit Metadata