claude-peers-mcp-inter-instance-messaging

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

• Third-party content exposure detected (medium risk: 0.30). The workflow reads peer summaries and messages from other local Claude Code instances via list_peers and message polling, which can contain untrusted free text from other local users or processes sharing the machine.

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

• Potentially malicious external URL detected (high risk: 0.70). The skill instructs cloning an external repository from GitHub (https://github.com/louislva/claude-peers-mcp.git) and executing its setup/runtime code via Bun, which constitutes an unofficial operational dependency with weak/unknown provenance.