The Agent Skills Directory

[SAFE]: The skill uses Model Context Protocol (MCP) integrations for Amplitude, Google Calendar, and Gmail to perform its tasks. These tools are used appropriately for the stated purpose of finding users and drafting outreach.
[SAFE]: Explicit safety constraints are built into the prompt template, specifically the rule that the agent must 'NOT send any email' and instead 'Save everything as drafts'. This ensures that a human remains in the loop to review data and communication before it is externalized.
[SAFE]: The skill does not exhibit any malicious patterns such as prompt injection, obfuscation, or unauthorized data exfiltration. All data flows (from Amplitude to Gmail) are transparent and aligned with the user's intent for research recruitment.

interview-scheduling