The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests and processes untrusted data from external web sources, creating a surface for indirect prompt injection.
Ingestion points: The scripts/search_client.py script retrieves search results and raw page content from the api.aisa.one API endpoints, which aggregate data from across the web.
Boundary markers: There are no explicit delimiters or instructions within the script or skill documentation to isolate external content from the agent's instructions.
Capability inventory: The skill allows for broad web searching and content extraction, which the agent uses to inform its decisions and responses.
Sanitization: The script performs basic truncation but does not sanitize or escape the content to prevent potential prompt injection payloads embedded in snippets or page text.
[SAFE]: The skill communicates with the vendor's official API infrastructure.
Evidence: The AISA_BASE URL in scripts/search_client.py points to https://api.aisa.one, which is the official domain for the aisa-team vendor, and API keys are correctly handled via environment variables.

web-search