Security audit
Security checks for vulnerabilities and agentic risk
This skill does what it claims, but it can send a Subscut API key to any configured HTTPS API endpoint, which needs review before installation.
Install only if you are comfortable sending the provided video URL and related processing request to Subscut. Keep SUBSCUT_API_BASE_URL at the default https://subscut.com unless you are deliberately testing with a separate non-production key, and do not let untrusted content choose --api-base-url. Rotate the API key if it may have been used with an untrusted endpoint.
scripts/generate-podcast-clips.js:57Configurable API Endpoint Can Receive the Bearer Credential
The skill declares environment and network capabilities but does not define an explicit tool scope such as permissions or allowed-tools. That creates ambiguity about what the runtime may permit, increasing the risk of overbroad access or unintended execution paths when the skill is invoked.
The skill does not clearly warn that the user-provided video URL and associated content will be sent to Subscut's external API for processing. This can cause unintentional disclosure of private or sensitive media, especially if users assume processing is local or do not realize third-party transfer occurs.
The script exports an API key value directly, which involves handling sensitive credentials, but there is no confirmation, warning, or explanatory note about protecting secrets or replacing the placeholder securely. In a code file, access to sensitive environment variables or credentials should include some visible disclosure unless clearly covered by the skill description or comments.
This JSON manifest describes the skill's purpose in broad natural language but does not specify concrete activation phrases, exclusion conditions, or negative examples. In manifest files, such ambiguity can cause the skill to be matched for general requests about clips, captions, or videos beyond the intended narrow use case.
This code performs an outbound HTTP POST to a remote service and includes both a bearer token and the user-supplied video URL in the request body. Although the help text documents required arguments and environment variables, the script does not explicitly warn the user at execution time or in comments/docstrings that media metadata will be transmitted to an external service.
No suspicious patterns detected.