Skills
Navigate
skills/didit-protocol/skills/didit-liveness-detection/Gen Agent Trust Hub

didit-liveness-detection

Pass

Audited by Gen Agent Trust Hub on Oct 13, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill includes a Python utility script (scripts/check_liveness.py) designed to interface with the Didit Passive Liveness API. It performs standard network operations using the requests library to upload images and metadata to the vendor's verified endpoint at https://verification.didit.me/v3/passive-liveness/.
  • [DATA_EXPOSURE]: The skill utilizes the DIDIT_API_KEY environment variable for authentication. This is a standard and recommended practice for secret management, preventing the hardcoding of credentials.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes external data including user images and metadata objects.
  • Ingestion points: Untrusted data enters via the user_image file and metadata JSON object parameters in scripts/check_liveness.py and SKILL.md examples.
  • Boundary markers: None explicitly defined for the prompt context, but parameters are passed to a structured API request.
  • Capability inventory: The skill is limited to performing a POST request to the Didit API and printing the response.
  • Sanitization: The script includes a validation step to ensure the --metadata argument is a valid JSON object before execution.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 13, 2026, 12:49 AM