AI Tool Launches

Turn spoken feedback into tasks your coding agent can run

The IDE that builds itself

See where your AI coding spend actually goes

Open Source harness for multi-agent coding workflows

Local semantic search for AI agents

Let Claude or Codex create and maintain good looking docs

Fine-tune an 8B LLM on a 4 GB laptop GPU

Build evals and custom benchmarks for real-world tasks

Open-source AI debugging platform that fixes production bugs

A terminal built for Claude Code with one-click session resume

Git-backed skills, tools & context for AI agents

Local sandboxes for AI agents on your Mac, Linux, bare metal

The Engineering Operating System for AI Coding Agents

Let AI agents put screenshots and videos on pull requests

Spend up to 85% less and run 3x longer coding agent sessions

Security scanner for agent skills — catches prompt injection before you install.

Self-evolve Gemini CLI instructions and skills — GA + GEPA/DSPy with hard gates before apply.

Open-source TUI coding agent — multi-provider routing, MCP, LSP, Plan/Agent/YOLO modes.

Peekaboo for Linux — screenshot, mouse, keyboard & window automation for desktop AI agents.

Security scanner for agent skills — catches prompt injection & supply-chain risks pre-install.

Never miss a Claude Code session — floating macOS window that surfaces when input is needed.

Code review studio for local git changes made by AI coding agents — Claude Code, Cursor, Copilot.

Runtime authorization layer for AI coding agents — Claude Code, Cursor, Codex, OpenClaw.

Instant, hardware-isolated sandboxes for AI agents.

Comment on your AI agent's code and collaborate on changes.

A macOS cockpit for working with Claude Code — terminal, browser, and file viewer in one.

6.4x faster than llama.cpp on Apple Silicon — local LLM inference, zero token cost.

Full-stack email built for AI agents — they sign up, send and reply on their own.

Shared memory across all apps and LLMs — persistent agent context in Claude.

Persistent memory and grounded knowledge for MCP agents — drop in any file or URL, get searchable kn

Full-stack email built for AI agents — they sign up, send, receive, and reply with zero human steps.

Private inference endpoint for coding agents — open models, EU infrastructure, zero data retention.

Private, fast inference API for coding agents — open-weight models, zero request retention, OpenAI-c

Your whole stack, running itself — an agent framework for autonomous operations.

Multiplayer Claude Code — pair program with agents, share live sessions.

Headless iOS app publishing pipeline for AI agents.

Open-source workspace for AI agents and workflows.

Give your AI agents design taste from existing websites.

Payments layer for agentic commerce: paywall APIs, MCP tools, and hosted endpoints.

OSS IDE for controlling AI coding agents with built-in loops

The CLI your coding agent uses to scaffold, eval, and ship AI agents.

OSS Git-native IDE for controlling AI coding agents with built-in loops.

Claude Code, Codex, any CLI agent — driven from your iPhone.

Compress tokens before your LLM calls — same context, ~50% cheaper.

Disposable cloud test environments for AI coding agents.