The Agent Skills Directory

[SAFE]: No malicious patterns or security risks were identified in the skill. The instructions are focused on SEO data analysis and generation.
[INDIRECT_PROMPT_INJECTION]: The skill handles untrusted data by analyzing page sources for structured data. 1. Ingestion points: Web page source code scanning for JSON-LD, Microdata, and RDFa (SKILL.md). 2. Boundary markers: None provided for external inputs. 3. Capability inventory: No file system writes, network requests (beyond data retrieval), or subprocess execution found in the instructions. 4. Sanitization: The skill recommends validating JSON-LD syntax and checking for placeholder text, though it does not explicitly address malicious injection in the data it processes.

seo-schema