Designing, evaluating, and shipping autonomous AI agents.
7650 posts
Human controlled project context
Agent Receipts / Obsigna plugin for OpenClaw — cryptographically signed audit trail for agent actions
AI Harness build on top of typescript
Residential proxy MCP for AI agents — bypass Cloudflare and anti-bot walls. npm: novada-proxy-mcp.
Autonomous DFIR agent driving SANS SIFT forensic tools via an MCP server + LangGraph, with guardrails and a hash-chained audit trail.
Safety first. Hardware-verified DeFi for AI agents — designed for when the AI can be compromised.
The agent eval standard for MCP — score output quality, catch safety failures, enforce cost budgets
An open-source, self-hosted API that turns standard email providers (Mailgun, SES, SendGrid) into "Inbox-as-a-Service" for AI Agents.
Persistent memory MCP server for Claude Code — self-hosted, n8n + PostgreSQL + pgvector. No cloud required.
gazebo-mcp — MCP server for Gazebo / gz-sim (mock + live bridge) for AI agents
The financial home AI agents run on.
MCP server for Proxmox VE: inventory and operate VMs, LXC containers, and nodes from an AI client, with a gated write-safety model.
