The Agent Skills Directory

[SAFE]: No malicious patterns or security risks were detected. The skill serves as a checklist and provides a functional code example for developers working with Persona identity verification.
[DATA_EXPOSURE_AND_EXFILTRATION]: The skill uses os.environ to access the PERSONA_API_KEY. This is a recommended security practice for managing secrets in environment variables rather than hardcoding credentials.
[INDIRECT_PROMPT_INJECTION]: The skill defines a data ingestion surface through API calls to Persona.
Ingestion points: The code snippet fetches inquiry data from https://withpersona.com/api/v1/inquiries (SKILL.md).
Boundary markers: None present in the provided Python snippet.
Capability inventory: The skill is configured with Read, Bash(curl:*), and Grep tools.
Sanitization: The script prints ID and status attributes directly from the JSON response.
Note: Because the data source is a well-known identity service, the risk of malicious instructions being embedded in these specific attributes is considered negligible.

persona-prod-checklist