# Orkas-AI/Orkas

Open-source multi-agent AI desktop client — build and command your AI agent team through conversation. A commander LLM dispatches sub-agents in parallel or in series; agents self-evolve via reflection and skill crystallization. Local-first, BYO LLM keys (Claude · OpenAI · Gemini · DeepSeek · Kimi · GLM · Qwen). macOS / Windows / Linux.

Repository: https://github.com/Orkas-AI/Orkas
Canonical: https://ross.abutalabs.com/products/orkas
Homepage: https://orkas.ai?source=github
Language: TypeScript
License: MIT
License Family: permissive
Topics: agent-collaboration, agent-orchestration, agent-team, ai-agents, ai-assistant, ai-team, ai-workflow, autonomous-agents, byo-llm, desktop-app, electron, llm, local-first, multi-agent, multi-agent-system, self-evolving-agents, agentic-ai, cross-platform, mcp, mcp-client
Last push: 2026-08-26T04:03:17+00:00

## Health v2 (maintenance only)
Score: 81/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 99, release rhythm 99, longevity 9
- inputs: {"age_days": 126, "days_push": 7, "days_rel": 8, "gap_med": 8.0, "n_releases_24m": 15}
- flags: young
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1453, forks 66 (observed 2026-08-28T04:04:45.680327+00:00)

## What it is
Orkas is an open-source, local-first multi-agent desktop application where a Commander LLM plans goals and dispatches specialist sub-agents in parallel or sequence. It ships with nine built-in agents (research, writing, slides, video, office files, coding, etc.), supports MCP, and lets users bring their own LLM keys from providers like Claude, OpenAI, Gemini, and DeepSeek.

## Use cases
- orchestrate a team of AI agents from a single chat
- research competitors and generate a report and slide deck automatically
- delegate parallel AI sub-tasks to specialist agents
- build documents, presentations, and videos with AI agents
- connect my own LLM API keys to a desktop AI assistant
- run MCP tools and external CLI agents in a multi-agent workflow
- automate multi-step business workflows with self-evolving agents

## When to choose
- you want one conversational interface to coordinate multiple specialized AI agents rather than juggling separate chatbots
- you need real file deliverables like reports, PPTX decks, videos, or office documents produced end-to-end
- you prefer local-first desktop software with bring-your-own model keys across many providers
- you want MCP support and a marketplace of ready-made agents and skills

## When to avoid
- you need a lightweight CLI or headless agent framework to embed in your own codebase
- you require fully offline operation with no LLM API usage or credits
- you want a simple single-chatbot assistant without multi-agent orchestration overhead
- you need guaranteed enterprise SLAs or mature long-term stability from a young project

## Facets
- artifact type: application
- maturity: active
- function: agent-framework, chat-interface, mcp, llm-inference, workflow-automation, chatbot
- domain: large-language-models, developer-tools
- platform: windows, cross-platform
- tags: multi-agent, agent-orchestration, commander-pattern, byo-llm, local-first, self-evolving-agents, ai-team, electron-app, mcp-client, agent-marketplace, ai-agents, automation, macos, linux, desktop, electron

## Member repositories
- Orkas-AI/Orkas (main) score 81

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:45.680327+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 2026-08-30T04:35:44.068692+00:00, confidence not recorded.
  - readme: https://github.com/Orkas-AI/Orkas (fetched 2026-08-28T04:04:45.680327+00:00, sha 730b56f0c7d1)
  - homepage: https://orkas.ai?source=github (fetched 2026-08-29T11:45:18.072467+00:00, sha b3272d66abac)
  - site_page: https://orkas.ai/docs (fetched 2026-08-29T11:45:18.083940+00:00, sha ba1e587ac3c5)
  - site_page: https://orkas.ai/pricing (fetched 2026-08-29T11:45:18.082057+00:00, sha f2ba446e9025)
- Data as of 2026-08-30T08:39:29.467469+00:00.