platform-security
Pass
Audited by Gen Agent Trust Hub on Sep 2, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides standard architectural guidelines for implementing authentication and authorization using a dedicated internal library (
platform-security). - [DATA_EXPOSURE_&_EXFILTRATION]: The skill explicitly advises against security anti-patterns: 'Never log tokens, refresh tokens, OTPs, passwords, private keys, or authorization headers' and 'Configure keys and secrets through secure configuration, not source code'.
- [REMOTE_CODE_EXECUTION]: No remote code execution patterns or external scripts were detected. The dependency provided (
com.bankfab.ksa.middleware:platform-security) appears to be a standard internal organizational artifact. - [INDIRECT_PROMPT_INJECTION]: While the skill defines how to process authentication context, it does not involve the ingestion of untrusted natural language data that could influence the agent's behavior. It follows a structured schema for token requests and validation.
- [CREDENTIALS_UNSAFE]: The configuration examples use placeholders (
/path/to/private-key.pem) and explicitly recommend using project secret management for sensitive material.
Audit Metadata