spec

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

• Third-party content exposure detected (high risk: 0.70). The skill may fetch a GitHub issue body at runtime (outsider-authored free text) when the input is a GitHub issue URL/identifier, and that issue body is then read and used as the upstream input to draft the spec (LLM ingests it as readable prose).

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

• Potentially malicious external URL detected (high risk: 0.90). The spec skill explicitly fetches a GitHub issue at runtime (e.g. https://github.com///issues/ or the short owner/repo#NNN form) to read the issue title/body as its upstream input, and that fetched content directly controls the agent's prompts/output, so it is a required runtime external dependency.