The Agent Skills Directory

[COMMAND_EXECUTION]: The skill workflow involves running a local script scripts/parse-sitemap.ts to parse sitemap files. This is consistent with the technical nature of the skill.
[DATA_EXFILTRATION]: The agent is instructed to fetch /sitemap.xml files from remote servers. This network activity is the primary function of the skill and does not target sensitive data.
[SAFE]: No obfuscation, credential misuse, or persistence mechanisms were found. The skill processes external data, posing a theoretical indirect prompt injection surface which is standard for web-parsing tools.
[PROMPT_INJECTION]: Potential surface for indirect prompt injection via external data ingestion.
Ingestion points: External sitemap files fetched in workflow step 1.
Boundary markers: None explicitly defined in the workflow instructions.
Capability inventory: Local script execution and URL cross-checking logic.
Sanitization: No specific sanitization or escaping steps are mentioned in the instruction markdown.

seo-sitemap