ping-orchestration-ios-davinci-sdk

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The prompt instructs the agent to solicit required parameters (notably the OAuth clientId) from the user and "substitute all placeholders" with those values in generated source files, forcing the LLM to include user-supplied credential-like values verbatim in its output.

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

• Third-party content exposure detected (high risk: 0.80). The skill's templates and docs (e.g., DavinciViewModel.swift.template, ContinueNodeView.swift.template and references/davinci-sdk.md) explicitly call davinci.start() and iterate ContinueNode.collectors returned from the PingOne DaVinci server, ingesting and acting on server-provided (third-party) collector data that can drive navigation and trigger actions, so untrusted remote content can materially influence behavior.