configuring-sso-and-scim

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The skill includes examples that place secrets verbatim into commands and cluster settings (e.g., client_secret, ldapbindpasswd, SCIM bearer token, certificate PEM placeholders and HBA entries), which would require the LLM to embed actual secret values in generated SQL/commands if applied—an exfiltration risk.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.80). The skill's required workflow instructs entering IdP metadata and OIDC discovery URLs and enabling automatic JWKS fetching (see Part 1.2a/1.2b and Part 3.2), which causes the system/agent to fetch and ingest untrusted, public IdP metadata/JWKS from third-party endpoints that can materially influence authentication/behavior.