AgentIndex icon
AgentIndex
ToolsMCP ServersHOTCategoriesTrendingNewBest ListsCompareGuides
Submit Tool
ToolsMCP ServersHOTCategoriesTrendingNewBest ListsCompare
Home/
Multi-Agent/
Orkas
Orkas logo

Orkas

Active·★ 2.2k·MIT·Updated 2026-10-22
★ Trending

Orkas is an open-source, local-first multi-agent desktop app. You describe a goal in a chat, and its Commander plans the work, handles general parts itself and coordinates specialist agents in parallel or in sequence. The README says nine specialists ship with the app out of 30 in a marketplace, including DeepResearcher for evidence-grounded research, ContentWriter for posts and reports, PptMaker for slide decks and ProductDeveloper for repo-aware engineering. Its worked example turns a request to research five competitors into a report and a deck in one chat, with finished files on your disk.

Beyond its own agents, Orkas can drive five external CLI agents as local subprocesses: the coding agents Claude Code, Codex and OpenCode, and the personal agents OpenClaw and Hermes-Agent, plus open-source tools such as HyperFrames. You bring your own model keys (Claude, OpenAI, Gemini, DeepSeek, Kimi, GLM, Qwen, MiniMax, Doubao) or point it at a local OpenAI- or Anthropic-compatible endpoint, and you can mix providers across agents. The README states that conversations, files, API keys, knowledge bases and custom agents stay on your disk, that model calls go from your machine to the provider rather than through Orkas servers, and that each agent keeps private skills and memory and improves through reflection.

© 2026 AgentIndex.app|Built by a 10-year iOS Developer.
Best by use caseAboutContactAdvertisePrivacy PolicyTerms of UseQYSGitHubBuy me a coffee ☕

Browse by Category

MCP ServersCode AssistantWorkflow AutomationRAG / Knowledge BaseMulti-AgentBrowser AutomationLLM InfraDev ToolingObservability

Not affiliated with Anthropic, OpenAI or Microsoft.

Guides

Packaged installers exist for macOS (Apple Silicon and Intel) and Windows; Linux runs from source (`./run.sh`, Node 22.12+, glibc 2.34+) and Alpine is not supported. The README notes that the packaged desktop app asks you to sign in on first launch, while the source build has no account layer, and that an optional built-in Orkas model is billed in credits whereas bring-your-own-keys use is free. It is MIT licensed and the last push was 2026-10-15. This review is based on the README and repository data; we have not run it ourselves.

This write-up is based on the project's README and public repository data. We have not hand-tested the tool. Data as of 2026-10-16.

License
MIT
Latest release
v2026.10.1 · 2026-10-01
Last commit
2026-10-16
Contributors
11
Forks
91
Open issues and PRs
10
First published
2026-04-29

Source: GitHub, npm and PyPI public APIs.

#agent-collaboration#agent-harness#agent-orchestration#agent-team#ai-assistant#ai-team#ai-workflow#autonomous-agents
$ Install
$ git clone https://github.com/Orkas-AI/Orkas.git && cd Orkas && ./run.sh
↗ Visit site★ GitHub
01

Features

01A Commander that plans a goal and dispatches it to specialist agents in parallel or in sequence; nine specialists ship with the app
02Drives external CLI agents (Claude Code, Codex, OpenCode, OpenClaw, Hermes-Agent) as local subprocesses from one chat
03Bring your own model keys: Claude, OpenAI, Gemini, DeepSeek, Kimi, GLM, Qwen and others, or a local OpenAI- or Anthropic-compatible endpoint
04Each agent keeps its own private skills and memory and improves through reflection after each task
05Local storage of conversations, files, knowledge bases and keys; macOS and Windows installers, Linux from source
02

Why choose it

+A Commander plus nine bundled specialist agents, so a single goal can become research, a report and a deck in one chat
+Can drive Claude Code, Codex, OpenCode, OpenClaw and Hermes-Agent as local sessions from the same chat
+Local-first: conversations, files, keys and knowledge bases stay on your disk, and model calls go straight to the provider
+Mix providers per agent, or use a local OpenAI- or Anthropic-compatible endpoint; MIT licensed
03

Trade-offs

!The packaged desktop app asks you to sign in on first launch, and the optional built-in Orkas model is billed in credits
!Packaged installers cover macOS and Windows only; Linux means running from source with Node 22.12+ and a large first-launch download
!A multi-agent Commander that can drive coding CLIs can edit files and run commands on your machine, so review permissions and what it is allowed to touch
!The project is young (repository created 2026-04-29), and the comparisons in the README are the project's own
05

Maintenance check

ActiveChecked: 2026-10-22
Last commit
2026-10-22
Commits, last 90 days
100+
Releases, last 90 days
8
Latest release
v2026.10.17 (2026-10-17)
Open issues
10
Recent issues with at least one comment
—
Commits per week, last 12 weeks
Commits per week, last 12 weeks: 0, 6, 14, 43, 10, 27, 7, 24, 32, 7, 34, 13

Oldest week on the left; the latest four weeks are orange. Source: GitHub commit activity.

Status rule: archived on GitHub = Archived; last commit within 90 days = Active; 91 to 365 days = Slowing; older = Stale. All figures come from the GitHub API on the date shown.

04

Who it fits

Pick it if
  • You want a desktop app where one chat coordinates several agents, instead of writing orchestration code.
  • You already use Claude Code, Codex or OpenCode and want one place to dispatch work to them.
  • You want your data, keys and agents on your own disk, with your own model keys.
Look elsewhere if
  • You only want a single all-purpose chatbot; the README itself says Orkas is not for that.
  • You want a code library to embed in your own app, or a fully hosted team: the README lists both as not a fit.
  • You need Linux packages today: it runs from source on glibc 2.34+ only.
06

Compatibility

Claude Code
Driven as a local CLI session
Verified via docs
Codex
Driven as a local CLI session
Verified via docs
OpenCode
Driven as a local CLI session
Verified via docs
07

Quick start

1
$ git clone https://github.com/Orkas-AI/Orkas.git
2
$ cd Orkas
3
$ ./run.sh
08

Use cases

↳Turn one goal into research, a written report and a slide deck through a chain of specialist agents
↳Break a product spec into tasks and hand them to coding agents such as Claude Code or Codex
↳Chat with local documents and run data analysis without sending files to a vendor cloud
09

Alternatives

⚡Top VS:
VSAgentCrew★224→VSp2pclaw-mcp-server★15→VSagent-fleet-o★76→

Comments

Log in to leave a comment

No comments yet. Be the first!

On this page
01Features02Why choose it03Trade-offs05Maintenance check04Who it fits06Compatibility
VS
systemprompt-template
★29
→
AgentCrew logo
AgentCrew★ 224
Chat application with multi-agents system supports multi-models and MCP
VSCompare→
p2pclaw-mcp-server logo
p2pclaw-mcp-server★ 15
MCP server + REST API for the P2PCLAW decentralized AI research network. Lets agents publish papers, vote on the mempool, and join the swarm. Powers www.p2pclaw.com.
VSCompare→
agent-fleet-o logo
agent-fleet-o★ 76
Open-source AI agent orchestration platform — self-hosted mission control for autonomous multi-agent systems. Visual DAG workflows, 450+ MCP tools, human-in-the-loop approvals. Works with Claude, GPT-4o, Gemini, Ollama, Codex.
VSCompare→
systemprompt-template logo
systemprompt-template★ 29
AI Governance Infrastructure — local evaluation. The governance layer for AI agents: a single compiled Rust binary that authenticates, authorises, rate-limits, logs, and costs every AI interaction. Self-hosted, air-gap capable, provider-agnostic.
VSCompare→
agentos logo
agentos★ 682
TypeScript AI agent framework: cognitive memory, runtime tool forging, multi-agent orchestration, 11 LLM providers.
VSCompare→
agency-orchestrator logo
agency-orchestrator★ 2.4k
One sentence → multiple AI roles collaborate → complete plan in minutes. 179 expert roles, zero-code YAML, 9 LLM providers. 一句话让多个AI角色自动协作，几分钟出完整方案。
VSCompare→
PilotDeck logo
PilotDeck★ 4.8k
Task-oriented AI Agent productivity platform
VSCompare→
multica logo
multica★ 53.6k
Make humans and AI agents work as one team — open-source and self-hostable.
VSCompare→
See all alternatives →
07
Quick start
08Use cases
09Alternatives
Orkas-AI /
Orkas
↗

MIT local-first desktop app where a Commander plans goals and coordinates specialist agents and CLI agents such as Claude Code and Codex

2.2k91MIT
Last commit1d ago
StatusActive
CategoryMulti-Agent
Contributors11
Latest releasev2026.10.1
Links
Documentation↗Discussion↗Issues↗Releases↗
Cloud Infrastructure
Ad

Deploy on DigitalOcean

Host your MCP servers, Docker containers, and AI agents with $200 free credit over 60 days.

Claim $200 Free Credit