zoho-desk

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

• Third-party content exposure detected (high risk: 0.75). At runtime, the skill uses ClawLink to access Zoho Desk via browser/OAuth and then the agent reads Zoho Desk content (e.g., ticket text/comments) that is authored by other users in the Zoho tenant, which is outsider free text injected into the LLM context through the Zoho Desk action results.

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

• Potentially malicious external URL detected (high risk: 1.00). The skill's runtime calls (e.g., npx @useclawlink/cli login / npx @useclawlink/cli connect zoho-desk) fetch and execute the @useclawlink/cli package from npm (see https://www.npmjs.com/package/@useclawlink/cli and https://claw-link.dev), so it relies on and executes remote code at runtime.