AI Infrastructure Articles

Memoria brings Git-level version control to AI agent memory with snapshots, branches, and rollback. Built on MatrixOne's Copy-on-Write engine, it eliminates hallucinations and enables safe experimentation.

Tokscale is a high-performance Rust-powered CLI tool that tracks token usage and costs across 20+ AI coding assistants including Claude Code, Cursor, Codex, and OpenCode. Features interactive TUI, real-time pricing, social leaderboards, and 10x faster native processing.

Discover OpenCode Studio, the local GUI that eliminates manual JSON editing for OpenCode configurations. Manage MCP servers, skills, plugins, and auth through an intuitive web interface with profile isolation and GitHub sync.

Claude Code Router lets you route AI coding requests to cheaper, faster models while keeping Anthropic's exceptional CLI interface. Cut costs by 90% with intelligent task-based routing to DeepSeek, Gemini, Ollama, and more.

LACP is a control-plane-grade agent harness for Claude, Codex & Hermes with policy gates, verification loops, 5-layer memory, and auditable execution. Learn how to harden your AI agent workflows in under 5 minutes.

VoidLLM is a privacy-first, self-hosted LLM proxy with zero-knowledge architecture, multi-provider routing, MCP gateway, and team governance. Deploy in seconds with Docker.

OpenClaw-bot-review is a lightweight, zero-database web dashboard for OpenClaw AI agents. Monitor bots, models, sessions & gateway health in real-time with delightful pixel-art visualization. Deploy in 5 minutes.

rtk is a high-performance Rust CLI proxy that reduces LLM token consumption by 60-90% on common dev commands. Single binary, zero dependencies, 100+ supported commands, and integrations with 13 AI coding tools including Claude Code, Copilot, and Cursor.

PinchTab is a 15MB Go binary that gives AI agents direct browser control via HTTP API. With token-efficient extraction (800 tokens/page), multi-instance orchestration, and local-first security, it slashes API costs by 5-13x compared to screenshot-based approaches.

Discover claude-usage, the zero-dependency Python dashboard that transforms hidden Claude Code logs into complete cost visibility. Track tokens, sessions, and spending across API, Pro, and Max plans with incremental scanning and real-time browser visualizations.

Discover how Zep's sub-200ms context engineering platform uses temporal knowledge graphs to transform AI agent reliability. Complete setup guide, real code examples, and architectural deep dive included.

agent-fetch is an open-source full-content web fetcher for AI agents that uses browser impersonation and 7 parallel extraction strategies to bypass anti-bot systems and retrieve complete articles as clean markdown.

Chorus is the open-source agent harness that orchestrates AI agent sessions with automatic recovery, human verification gates, and multi-agent collaboration. Deploy in 2 commands with embedded PostgreSQL.

NullClaw is a revolutionary Zig-based AI assistant infrastructure featuring a 678KB binary, ~1MB RAM usage, and sub-2ms startup. Perfect for edge computing and microcontrollers, it supports 50+ providers and 19 channels without compromising on performance.