omnicaptions-transcribe

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.80). The skill prompt instructs the agent to ask the user for their API key and pass it via the command-line argument -k <key>, which causes the secret to be handled and output directly by the LLM context.

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

• Third-party content exposure detected (low risk: 0.10). The required workflow processes user-specified media URLs (such as YouTube videos) and local files, which falls under fetching selected external URLs or web content (low risk).

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.70). The setup installs a package from an arbitrary non-official GitHub Pages PyPI repository (https://lattifai.github.io/pypi/simple/), constituting a weak/unknown operational dependency.