Show HN: I built Comrade – the security-focused AI agent

Category: ai-ml

Tags: ai-agent, local-first, mcp-integration, automation, security

Score: 6.8/10 (Innovation: 6, Technical: 7, Documentation: 8, Utility: 6)

Comrade is a security-focused, local-first AI agent workspace that provides a comprehensive suite of over 34 tools for file operations, shell execution, web automation, and MCP integration. It's interesting because it combines extensible agentic workflows with a strong emphasis on transparency and data privacy, positioning itself as a secure alternative to cloud-based AI assistants. The project integrates Model Context Protocol for external tool connectivity while maintaining local execution by default.

Target audience: devops, backend devs, security engineers, ai developers

Repository: https://github.com/LaurentiuGabriel/comrade · TypeScript · MIT

View on Hacker News