seo-article-review
Pass
Audited by Gen Agent Trust Hub on Jun 20, 2026
Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes a local script
scripts/run.pyto transmit diagnostic parameters and article content to the AI Skills API. - [DATA_EXFILTRATION]: User-provided content, including article text, uploaded files, and provided URLs, is sent to the domain
https://ai-skills.ai. This is the intended behavior for the service to perform its analysis. - [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it ingests and processes untrusted external content (Category 8).
- Ingestion points: Untrusted data enters via the
materialText,materialFile, andmaterialUrlparameters inSKILL.md. - Boundary markers: The skill lacks explicit boundary markers or instructions to the AI to ignore instructions embedded within the user-provided articles.
- Capability inventory: The agent has the capability to execute network requests via the
scripts/run.pywrapper to interact with the service provider. - Sanitization: There is no evidence of content sanitization or filtering of the article data before it is processed by the diagnostic engine.
Audit Metadata