llm-space

Overview of llm-space

What it is

LLM Space is a desktop application for building and debugging AI agents. It allows users to prototype agent ideas, inspect every step of harness execution, debug failures, and evaluate performance. It is a sister project of DeerFlow and is used to build and debug every version of DeerFlow. The project started in March 2023, and v4 is its fourth major iteration. It is local-first, with files and API keys staying on the user's local computer. It collects a small amount of anonymous usage data (see TELEMETRY.md for details and opt-out).

Key features

• Build: Write and version prompts, system messages, tools, and model settings.
• Trace: See every model call and tool run inside the agent loop as it happens.
• Debug: Replay a run from history and step through it to find what went wrong.
• Evaluate: Measure how your agent performs across runs.
• Manage: Keep threads organized as files on your own machine.
• Generate: Let AI write your prompt and tools for you, and turn any thread into a runnable LangGraph agent.

Use cases

• Prototyping new agent ideas.
• Inspecting every step of harness execution.
• Debugging failures by replaying and stepping through runs.
• Evaluating agent performance across multiple runs.
• Managing and organizing threads locally.
• Generating prompts, tools, and runnable agents with AI assistance.

Getting started

Install from source (requires Bun): bash bun install Start the desktop app for local development: bash mise run dev Build a canary release: bash mise run build:canary Alternatively, download a DMG from the latest release for macOS (Apple Silicon and Intel). Two editions are available: LLM Space (system WebView, ~27 MB) and LLM Space Performance (embedded rendering engine, ~130 MB).

License and pricing

LLM Space is released under the MIT License. It is open source and free to use.

Based on the project's public README and repository metadata.