We build AI systems that can reason We aim to rekindle the dream of the personal computer—by creating practical AI agents that can accomplish larger goals and safely work for us in the real world.
Bouncer: Filters and heals your X (Twitter) feed.
Sculptor: An extensible, open-source app for running multiple coding agents in parallel with isolated workspaces.
Vet: A local code review tool that catches mistakes made by coding agents, providing high-signal output and CI-ready reviews.
Blueprint: A planning tool that generates coding plans through multiple-choice questions based on your codebase.
Latchkey: A credential management tool that allows local AI agents to access APIs securely without exposing sensitive information.
Bouncer: Filters and heals your X (Twitter) feed to improve content quality.
Sculptor: Provides an extensible, parallel coding workspace for running multiple coding agents simultaneously.
Vet: Acts as a local code review tool that catches mistakes made by coding agents, ensuring code quality.
Blueprint: Assists in planning coding tasks by generating structured plans through interactive multiple-choice questions.
Latchkey: Manages API credentials locally for AI agents, allowing secure access without exposing sensitive information.
Provides a suite of AI tools designed to enhance productivity and creativity for developers.
Offers open-source applications, ensuring users have control over their tools and can customize them as needed.
Facilitates parallel coding and agent management, allowing multiple tasks to be handled simultaneously without interference.
Ensures data privacy by keeping user code and credentials local, preventing external access or telemetry.
Supports a collaborative environment through community engagement and shared resources, fostering innovation and improvement.
Sculptor: Free, open source, and requires a paid Claude subscription for model usage.
Vet: Free for any use, including commercial, under AGPL-3.0 license.
Blueprint: Free and open source, MIT licensed.
Latchkey: Free and open source, with no additional costs for using supported services.
Keystone: Costs $1.00 USD of inference per run, with the ability to adjust the budget.
