The Agent Skills Directory

[SAFE]: The skill implements standard API connector functionality for DataForSEO. It uses the official api.dataforseo.com endpoint for all network operations.
[SAFE]: Authentication is handled securely through either Zapier-managed connections or environment variables (DATAFORSEO_LOGIN and DATAFORSEO_PASSWORD), avoiding hardcoded credentials.
[SAFE]: The execution environment is well-defined, using a standard cli.js wrapper to manage dependencies and runtime version checks.
[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from external sources such as Google Search results and AI model responses (ChatGPT, Claude, Gemini). While this represents an indirect prompt injection surface, it is the primary intended purpose of the SEO and AI-mentions tools. The implementation uses structured Zod schemas for all inputs and outputs to enforce data boundaries.
Ingestion points: API responses from api.dataforseo.com processed in lib/dataforseo.ts and returned by all scripts.
Boundary markers: Zod schemas are used for structural validation of API responses.
Capability inventory: Performs network requests to the vendor's API endpoint.
Sanitization: Standard structural validation is present; content-level sanitization for embedded instructions in fetched data is not implemented as it would interfere with the tool's purpose (retrieving raw SERP/LLM data).

dataforseo