unblocked-context-get-urls

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

• Third-party content exposure detected (high risk: 0.90). The SKILL.md explicitly allows calling context_get_urls to fetch full content from "any URL reachable through Unblocked's configured connectors ... and arbitrary public web pages" (examples: GitHub PRs, Jira, Slack threads), which the agent is expected to read and act on, exposing it to untrusted third‑party/user‑generated content that could carry indirect prompt injections.

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

• Potentially malicious external URL detected (high risk: 0.80). Yes — the skill's context_get_urls fetches the full body of arbitrary URLs at runtime (e.g., https://github.com/acme/payments-service/pull/89) and injects that content into the agent context, so remote content can directly control prompts.