The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to process external search results from SerpAPI, which are untrusted data sources that could contain malicious instructions.
Ingestion points: Search results retrieved via WebFetch or the search archive endpoint as described in the instructions.
Boundary markers: The skill instructs to "validate and normalize once" and suggests using "sanitized state-machine fixtures," though it does not mandate specific prompt delimiters.
Capability inventory: The skill is permitted to use Write and Edit for state management and WebFetch for network operations.
Sanitization: The instructions emphasize normalizing business fields rather than processing unstable raw payloads.
[EXTERNAL_DOWNLOADS]: The skill references official SerpAPI documentation and client repositories on GitHub.
Evidence: Links provided to serpapi.com and GitHub repositories under the serpapi organization (serpapi-python, serpapi-javascript). These are well-known resources for the stated functionality.
[DATA_EXPOSURE]: The skill provides instructions for managing the SERPAPI_KEY and explicitly warns against insecure practices.
Evidence: Instructions state "never key-bearing archive URLs" and suggest using environment variables for the server-side submitter.

serpapi-webhooks-events