skills/thelobbi/claude/keycloak-admin/Gen Agent Trust Hub

keycloak-admin

Pass

Audited by Gen Agent Trust Hub on Aug 25, 2026

Risk Level: SAFECREDENTIALS_UNSAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [CREDENTIALS_UNSAFE]: The documentation contains hardcoded default credentials in example curl commands for obtaining tokens (username=admin, password=admin) and setting temporary user passwords (temp_password_123). While these are presented as local development defaults, they represent hardcoded secrets in the instruction set.
  • [COMMAND_EXECUTION]: The skill involves executing host-level commands, specifically docker cp and docker restart, to manage identity provider services and deploy custom themes, which requires elevated environment permissions.
  • [PROMPT_INJECTION]: The skill facilitates the ingestion of user-controlled data into administrative API requests, creating a surface for indirect prompt injection. 1. Ingestion points: Fields from req.body (email, firstName, lastName, orgId) in the user-service controller are used to construct user objects. 2. Boundary markers (absent): No delimiters or instructions to ignore embedded commands are present in the interpolation logic. 3. Capability inventory: Comprehensive identity management including user creation, role assignment, and realm configuration via axios and curl. 4. Sanitization (absent): No input validation or sanitization logic is evidenced in the provided skill scripts.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 25, 2026, 09:07 AM
Security Audit — agent-trust-hub — keycloak-admin