affinity

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

• Third-party content exposure detected (medium risk: 0.65). At runtime, the skill’s shell-driven Affinity actions (via ClawLink) fetch and return Affinity records that can include outsider-authored free text fields (e.g., list entry “author”/notes or other user-entered text) into the agent’s LLM context through the CLI’s command output.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill instructs the agent to run npx @useclawlink/cli (which fetches and executes remote code from the npm package listed at https://www.npmjs.com/package/@useclawlink/cli) at runtime and the CLI is a required dependency for the skill’s functionality, so remote code execution is occurring.