mission-control

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.80). The skill instructs the agent to create a JSON config file containing fields like "hookToken", "botToken" and explicitly mentions creating the config with your GitHub token (and shows example token placeholders), which can lead the agent to request and embed secret values verbatim rather than using secure auth flows or environment variables.

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

• Third-party content exposure detected (high risk: 0.90). The skill's webhook transform explicitly fetches and loads task data from a GitHub repository (via the Git Blob API and the GitHub Pages dashboard) and the agent is instructed to "read task details" and "execute" work orders when tasks move to in_progress, so untrusted user-generated repo content can directly drive agent actions and tool use.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill's webhook transform fetches the exact tasks.json at runtime via the Git Blob API (e.g. GET https://api.github.com/repos/{owner}/{repo}/git/blobs/{blob_sha}) and injects that file's task descriptions into the agent's work-order/context, so remote content directly controls agent prompts/behavior.