The Agent Skills Directory

[SAFE]: The skill provides legitimate instructions for configuring Fathom AI API access using official service endpoints (api.fathom.ai and fathom.video).
[SAFE]: The provided code examples use environment variable placeholders (your-api-key-here) rather than hardcoded credentials, following security best practices.
[SAFE]: Explicit instructions are included to manage secrets in .env files and to update .gitignore to prevent credential exposure in repository history.
[INDIRECT_PROMPT_INJECTION]: The skill configures an attack surface for processing meeting transcripts and summaries which could contain untrusted data.
Ingestion points: External meeting data retrieved via the meetings API endpoint in SKILL.md.
Boundary markers: None present in the connection test examples to separate data from system instructions.
Capability inventory: Uses Bash(curl:*) for data fetching and Write/Edit for project configuration as specified in SKILL.md.
Sanitization: No specific content sanitization is implemented in these setup instructions.

fathom-install-auth