The Agent Skills Directory

[CREDENTIALS_UNSAFE]: The skill manages DataForSEO API credentials by storing them in a local .env file at ~/.claude/skills/seo/.env. The instructions explicitly direct the agent to set file permissions to chmod 600 to restrict access and warn against echoing credentials in outputs or committing them to version control. This follows established security practices for local secret management.
[EXTERNAL_DOWNLOADS]: The skill communicates with api.dataforseo.com to fetch SEO metrics, keyword data, and search engine results. This is an interaction with a well-known commercial service required for the skill's primary functionality.
[DATA_EXFILTRATION]: User-provided API credentials (login and password) are transmitted to api.dataforseo.com via Basic Authentication over HTTPS. This communication is directed to the official API endpoint of the service provider and is necessary for authorized data retrieval.
[PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface as it ingests untrusted data from external sources.
Ingestion points: Search engine results (titles and descriptions) fetched in prospect_finder.py and serp_check.py, as well as website metadata and content crawled in on_page_audit.py.
Boundary markers: The instructions in SKILL.md do not define specific delimiters or "ignore" instructions for the external content when it is passed to the agent for synthesis.
Capability inventory: The skill has access to Bash, WebFetch, Write, and Agent tools.
Sanitization: There is no evidence of explicit sanitization or filtering of the retrieved web content before it is processed by the agent.

seo