setup-api-key

Pass

Audited by Gen Agent Trust Hub on Sep 2, 2026

Risk Level: SAFEDATA_EXFILTRATIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [DATA_EXFILTRATION]: The skill reads the ElevenLabs API key from local environment variables or .env files and sends it to an external server for validation.
  • The network request is directed to api.elevenlabs.io, which is the official API endpoint for the skill's author.
  • [EXTERNAL_DOWNLOADS]: Performs a network GET request to https://api.elevenlabs.io/v1/user to fetch user account details and confirm the validity of the API key.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests data from local files, creating a potential surface for indirect instructions if the file content is compromised.
  • Ingestion points: Reads content from the .env file located in the user's project directory (SKILL.md).
  • Boundary markers: None are used during the file reading process; the skill searches for specific variable patterns.
  • Capability inventory: The skill can perform network operations and read local files.
  • Sanitization: The skill includes robust behavioral instructions to redact keys if mentioned and strictly forbids asking the user to paste keys into the chat window.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 2, 2026, 04:37 AM
Security Audit — agent-trust-hub — setup-api-key