The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill configuration connects to a remote MCP server hosted at https://nothumansearch.ai/mcp. This is required for the skill's primary functionality but establishes a dependency on the availability and integrity of the external service.
[DATA_EXFILTRATION]: The register_monitor tool is designed to collect and transmit a user's email address to the nothumansearch.ai infrastructure for the purpose of domain monitoring. While the skill instructions specify that this should only be done with explicit user consent, it represents a path for PII (Personally Identifiable Information) collection.
[PROMPT_INJECTION]: The skill processes data from a remote search index and inspects third-party websites (via search_agents, get_site_details, and verify_mcp). This introduces a surface for indirect prompt injection, as data retrieved from these external sources could contain malicious instructions designed to influence the agent's subsequent actions.
Ingestion points: Results from search_agents, get_site_details, get_stats, and verify_mcp (SKILL.md).
Boundary markers: None identified in the prompt templates.
Capability inventory: No local execution capabilities are defined in the skill itself; however, the tool is used to discover other services which the agent may subsequently interact with.
Sanitization: No explicit sanitization or validation of the remote search results is documented within the skill instructions.

not-human-search-mcp