skills/api7/a6/a6-plugin-jwt-auth/Gen Agent Trust Hub

a6-plugin-jwt-auth

Pass

Audited by Gen Agent Trust Hub on Sep 9, 2026

Risk Level: SAFECOMMAND_EXECUTIONCREDENTIALS_UNSAFEDATA_EXFILTRATION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes standard administrative commands to manage the Apache APISIX gateway. This includes using the a6 CLI for creating and updating routes, consumers, and credentials, as well as openssl for generating RSA key pairs for asymmetric authentication. These operations are consistent with the skill's stated purpose of plugin configuration.
  • [CREDENTIALS_UNSAFE]: No hardcoded secrets or sensitive credentials were found. The skill demonstrates secure practices by providing placeholders for keys and explicitly documenting how to use environment variables ($env://) and HashiCorp Vault ($secret://) to manage JWT secrets securely.
  • [DATA_EXFILTRATION]: Network activity is limited to local testing via curl to 127.0.0.1 and interactions with defined secret management endpoints. There are no patterns suggesting the exfiltration of sensitive data to unauthorized external domains.
  • [SAFE]: The skill is authored by the Apache APISIX Contributors, which is a trusted community, and it manages resources associated with the APISIX ecosystem (e.g., the a6 CLI). All operations and configurations described are legitimate administrative tasks with no evidence of malicious intent.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 9, 2026, 12:28 PM
Security Audit — agent-trust-hub — a6-plugin-jwt-auth