The Agent Skills Directory

[SAFE]: The skill utilizes official Didit domains (didit.me) for all API operations. These domains belong to the verified author 'didit-protocol', and the communication is conducted over secure HTTPS connections.
[CREDENTIALS_UNSAFE]: No hardcoded credentials were found. The skill correctly instructs the use of environment variables (DIDIT_API_KEY) for managing sensitive API keys, adhering to standard security practices.
[COMMAND_EXECUTION]: The provided Python script scripts/screen_aml.py uses the argparse module to handle user inputs as data parameters. No dangerous shell execution or unsanitized command injection vectors were detected.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to process user-provided identity information which could theoretically contain malicious instructions.
Ingestion points: Identity fields such as full_name and document_number passed to scripts/screen_aml.py.
Boundary markers: Absent; data is passed directly to the vendor API.
Capability inventory: The script is limited to performing network POST requests to the Didit API endpoint.
Sanitization: The script serializes inputs into a JSON payload, preventing structural injection into the transmission protocol.

didit-aml-screening