orca-cli

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The skill's worked example instructs the agent to fill a login form with a plaintext password ("s3cret") via an orca command, which would require the LLM to include secret values verbatim in its generated commands/outputs.

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

• Third-party content exposure detected (high risk: 0.78). The skill’s workflow uses the orca CLI (e.g., orca goto --url, orca snapshot --json, orca exec --command "get text/html/value/get title/get url") which at runtime reads publicly fetched page content and accessibility/text into JSON output that the agent will place into its LLM context, creating an indirect prompt-injection path from outsider-authored web/free-form page text.