# deer-flow/llm-space

A desktop app to prototype agent ideas, inspect every harness step, replay failures, and evaluate performance, all in one place. Local-first, cloud-ready for managed agents.

Repository: https://github.com/deer-flow/llm-space
Canonical: https://ross.abutalabs.com/products/llm-space
Homepage: https://deer-flow.github.io/llm-space/
Language: TypeScript
License: MIT
License Family: permissive
Topics: agent, deer-flow, develop, electrobun, harness, llm, llm-tools, managed-agents
Last push: 2026-08-22T05:19:47+00:00

## Health v2 (maintenance only)
Score: 80/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 99, release rhythm 98, longevity 4
- inputs: {"age_days": 66, "days_push": 11, "days_rel": 14, "gap_med": 1, "n_releases_24m": 30}
- flags: young
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1712, forks 181 (observed 2026-08-28T04:05:25.630510+00:00)

## What it is
LLM Space is a local-first desktop application for building, tracing, debugging, and evaluating LLM agents. It lets developers prototype agent ideas, inspect every harness step, replay failures, and measure agent performance in one place.

## Use cases
- prototype new LLM agent ideas
- inspect every step of agent harness execution
- replay failed agent runs to find bugs
- evaluate agent performance across runs
- manage agent threads as local files
- turn a conversation thread into a runnable LangGraph agent

## When to choose
- you are building or debugging LLM agents and want full visibility into each model call and tool run
- you want a local-first tool where prompts, API keys, and threads stay on your machine
- you need to replay and step through past agent runs to diagnose failures
- you want to benchmark and evaluate agent behavior across multiple runs

## When to avoid
- you need a production agent deployment platform rather than a development and debugging environment
- you require a fully web-based or team-hosted solution with no desktop app
- you are not working with LLM agents or harness-based workflows

## Facets
- artifact type: application
- maturity: active
- function: agent-framework, llm-inference, developer-tools, prompt-engineering, testing
- domain: large-language-models, developer-tools
- platform: cross-platform
- tags: agent-debugging, trace-inspection, local-first, llm-harness, agent-evaluation, electrobun, ai-agents, desktop

## Member repositories
- deer-flow/llm-space (main) score 80

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:25.630510+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T03:34:52.177022+00:00, confidence not recorded.
  - readme: https://github.com/deer-flow/llm-space (fetched 2026-08-28T04:05:25.630510+00:00, sha 2614af9173a8)
  - homepage: https://deer-flow.github.io/llm-space/ (fetched 2026-08-29T11:11:16.039804+00:00, sha 3b7a46727e2c)
- Data as of 2026-08-30T08:39:29.467469+00:00.