The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes local Python scripts scripts/seo_analyzer.py and scripts/generate_sitemap.py to perform its core functions. These scripts are deterministic and limited to analyzing local HTML files and writing sitemap XML files, which is consistent with the stated SEO purpose.
[SAFE]: Analysis of the source code and documentation found no evidence of prompt injection, data exfiltration, or obfuscation. External links included in the guides point to reputable SEO and validation services from Google and Schema.org.
[PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface because it processes untrusted HTML data provided by the user. 1. Ingestion points: scripts/seo_analyzer.py (reading HTML file content). 2. Boundary markers: Absent in the extracted analysis report. 3. Capability inventory: No dangerous capabilities such as eval(), exec(), or unauthorized network access are present. 4. Sanitization: Uses the standard Python html.parser library to extract specific tags, minimizing the risk of instructions influencing the agent.

seo-optimizer