web-browsing-cli

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

• Third-party content exposure detected (low risk: 0.10). In SKILL.md the web-browsing-cli workflow explicitly fetches and extracts text from a user-supplied URL via open <url>/do <n>/raw [url], and this extracted page text is LLM-ingested free text that an outsider can author on the underlying site.

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

• Potentially malicious external URL detected (high risk: 0.80). The skill invokes npx to fetch and execute the remote package @only-cli/oc@0.2.0 at runtime (e.g., "npx --yes @only-cli/oc@0.2.0 open "), which fetches and runs remote code that the skill relies on.