The Agent Skills Directory

[SAFE]: The skill implements a network whitelist that restricts all API requests to the official vendor endpoint https://api.octen.ai/search, effectively preventing unauthorized data exfiltration to third-party or malicious domains.\n- [SAFE]: Authentication is handled via the OCTEN_API_KEY environment variable, ensuring that sensitive API credentials are not hardcoded or exposed within the skill files.\n- [PROMPT_INJECTION]: The skill processes untrusted data from the live web, creating a potential surface for indirect prompt injection.\n
Ingestion points: The scripts/search.py script retrieves and displays external web content, including titles and highlights, from the Octen search index.\n
Boundary markers: Search results are presented in a markdown format without explicit isolation markers or security warnings to the agent context.\n
Capability inventory: The skill leverages network access via the urllib.request library and provides structured output to the agent.\n
Sanitization: There is no evidence of sanitization or filtering of the retrieved web content before it is passed to the agent.

OctenSearch