The Agent Skills Directory

[SAFE]: The skill utilizes official API endpoints for well-known services (Supabase, Apify, and Anthropic) to perform data storage, enrichment, and analysis.\n- [SAFE]: Environment variables for API keys are managed appropriately and are used solely for authentication with the respective official service providers.\n- [SAFE]: While the skill ingests external data from LinkedIn posts for LLM-based relevance scoring, the potential for indirect prompt injection is mitigated by the prompt structure and the restriction of the LLM output to a numeric integer used for internal logic.\n- [SAFE]: The skill implements a dry-run mechanism and includes explicit instructions for the AI agent to seek user approval before performing any database write operations, ensuring human-in-the-loop control.

signal-scanner