The Agent Skills Directory

[DATA_EXFILTRATION]: The skill connects to https://api.keenable.ai/mcp to perform web searches and fetch content. This domain is managed by the skill author and aligns with the skill's primary function.
[CREDENTIALS_UNSAFE]: The skill references the KEENABLE_API_KEY environment variable for rate limiting. It explicitly includes a security guideline for the agent to never ask users to provide or paste API keys into the chat interface.
[PROMPT_INJECTION]: The skill processes untrusted external data from the web using fetch_page_content and search_web_pages (SKILL.md). While this represents a surface for indirect prompt injection, no malicious exploitation patterns were detected.
Ingestion points: Web content retrieved via fetch_page_content and search_web_pages.
Boundary markers: None present in the skill instructions.
Capability inventory: Limited to web search and content retrieval.
Sanitization: No explicit sanitization or filtering of external content is defined in the instructions.

keenable-web