Orkas Orkas
Download GitHub
More use cases: Deep research Chat with documents Make videos SEO & GEO Write code Build apps
Home Use cases For Developers
</> For Developers

An AI agent team that
works in your repo.
Uma equipe de agentes de IA que
funciona no seu repositório.
一支在你代码仓库里
干活的 AI agent 团队。
あなたのリポジトリで
働く AI エージェントチーム。

Point Orkas at your codebase and hand off the grind: a lead agent recruits an engineer to reproduce the bug, write the fix, run your tests, and open the pull request — local-first, your own model keys, never proxied.Aponte Orkas para sua base de código e deixe a tarefa árdua: um agente líder recruta um engenheiro para reproduzir o bug, escrever a correção, executar seus testes e abrir a pull request - local-first, suas próprias chaves de modelo, nunca passa por proxy. 把 Orkas 指向你的代码库,把繁琐的活儿交出去:主 agent 召集一位工程师 agent 复现 bug、写修复、跑你的测试、提交 pull request——本地优先,使用你自己的模型密钥,绝不中转。 Orkas をあなたのコードベースに向けて、面倒な作業を任せましょう。リードエージェントがエンジニアを招集し、バグを再現し、修正を書き、テストを実行し、プルリクエストを開きます——ローカルファースト、あなた自身のモデルキーで、決してプロキシしません。

Runs on your machineReads your local repoBYO LLM · never proxied
What you can hand off

Developer workflows, done as a team

Each spans skills a single coding assistant would juggle clumsily — Orkas splits them across agents in one conversation.

Bug repro → patch → PR

The team reproduces the failure, writes the fix, runs the tests, and opens a pull request with the diff explained.

  • code_exec
  • file_io
  • git

Codebase Q&A from a local KB

Ingest the repo and docs into a local knowledge base; ask architecture questions and get cited, file-aware answers.

  • kb_ingest
  • kb_search

Interactive tools from a prompt

Generate a dashboard, log viewer, or config visualizer as a sandboxed app — saved to My Apps for reuse.

  • create_artifact
  • preview

Dependency & release triage

Scan open PRs and dependency updates from GitHub, summarize risk, and draft the changelog.

  • github
  • summarize
In one conversation

From “repro this” to a green PR

You describe the bug; the team does the loop. Risky steps pause for your approval before anything touches the repo.

Approval gates on writes & commands
Engineer team · 2 agents
Reproduce the launcher deadlock and open a PR.
L
Lead agent
Bringing in @engineer to reproduce and fix.
E
Engineer
Reproduced the hang, wrote the patch, tests green. skill: bash
E
Engineer
Opened PR #318 — fix(launcher): await lock release. connector: GitHub
How it works

Four moves, no config

01

Point at your repo

Set the working directory; the team reads your local code.

02

Describe the task

“Fix this bug”, “add tests”, “triage these PRs”.

03

Approve the plan

Review what it intends to run before it touches anything.

04

Get the PR

Diff explained, tests green, ready for review.

Safe on your codebase

Your code stays on your machine by default

Local-first

The repo, configs, and KB stay on disk — only the model API call goes out.

Approval gates

Writes, deletes, and shell commands pause for your sign-off.

BYO keys, never proxied

Use your own OpenAI / Claude / Gemini key; traffic goes direct.

See security →
FAQ

Developer questions

Can an AI agent fix bugs in my repository?Um agente de IA pode corrigir bugs no meu repositório? AI agent 能修复我代码仓库里的 bug 吗? AI エージェントは私のリポジトリのバグを修正できますか?

Yes. In Orkas you point an engineer agent at your local repo; it runs commands to reproduce the bug, writes a fix, runs your tests, and opens a pull request with the diff explained — all on your machine, using your own LLM key.Sim. No Orkas você aponta um agente engenheiro para seu repositório local; ele executa comandos para reproduzir o bug, escreve uma correção, executa seus testes e abre uma pull request com a diferença explicada - tudo em sua máquina, usando sua própria chave LLM. 可以。在 Orkas 里,你把一位工程师 agent 指向本地仓库;它运行命令复现 bug、写出修复、跑你的测试,并提交一个附带 diff 说明的 pull request——全程在你的机器上,使用你自己的 LLM 密钥。 はい。Orkas ではエンジニアエージェントをローカルリポジトリに向けます。コマンドを実行してバグを再現し、修正を書き、テストを実行し、差分の説明付きでプルリクエストを開きます——すべてあなたのマシン上で、あなた自身の LLM キーを使って行われます。

Is Orkas safe to run on my codebase?O Orkas é seguro para execução na minha base de código? 在我的代码库上运行 Orkas 安全吗? Orkas を自分のコードベースで実行しても安全ですか?

Orkas is local-first and open source, and risky steps like writing files or running commands pause for your approval. By default your code stays on your machine, and your keys and model traffic never go through Orkas — only the API call to your own provider leaves, and it's never proxied. If you turn on Lite's optional multi-device sync, the data you sync is stored on Orkas servers so it's available across devices; the free edition also sends limited usage analytics (usage metadata, not your content).Orkas é local-first e open source, e etapas arriscadas, como gravar arquivos ou executar comandos, são pausadas para sua aprovação. Por padrão, seu código permanece em sua máquina e suas chaves e tráfego de modelo nunca passam por Orkas - apenas a chamada de API para seu próprio provedor sai e nunca passa por proxy. Se você ativar a sincronização opcional de vários dispositivos do Lite, os dados sincronizados serão armazenados nos servidores Orkas para que fiquem disponíveis em todos os dispositivos; a edição gratuita também envia análises de uso limitadas (metadados de uso, não seu conteúdo). Orkas 本地优先且开源,写文件、跑命令等有风险的步骤都会暂停,等你批准。默认情况下,你的代码都在本机,模型 Key 与模型流量始终不经过 Orkas——只有发往你自己模型提供商的 API 调用会发出,且绝不被代理。若开启 Lite 的多端同步,同步的数据会存到 Orkas 以便跨设备使用;基础版还会采集有限的使用数据(使用元数据,并非你的内容)。 Orkas はローカルファーストでオープンソースで、ファイルの書き込みやコマンド実行などリスクのある操作は承認を待って一時停止します。既定では、あなたのコードはマシン上にとどまり、キーとモデルのトラフィックは Orkas を経由しません——送出されるのは自分のプロバイダーへの API 呼び出しだけで、決してプロキシされません。Lite の任意の複数端末同期を有効にすると、同期したデータは複数端末で使えるよう Orkas のサーバーに保存されます。無料版では限定的な利用解析(コンテンツではなく利用メタデータ)も送信されます。

Which models can the developer agents use?Quais modelos os agentes desenvolvedores podem usar? 开发者 agent 可以使用哪些模型? 開発者エージェントはどのモデルを使えますか?

You bring your own keys — OpenAI, Anthropic Claude, Google Gemini, and more — via OAuth or API key. Traffic is never proxied through Orkas servers.Você traz suas próprias chaves – OpenAI, Anthropic Claude, Google Gemini e outros – por OAuth ou chave de API. O tráfego nunca passa pelos servidores do Orkas como proxy. 你自带密钥——OpenAI、Anthropic Claude、Google Gemini 等——通过 OAuth 或 API key 接入。流量绝不经 Orkas 服务器中转。 OpenAI、Anthropic Claude、Google Gemini など、あなた自身のキーを OAuth または API キーで持ち込みます。トラフィックが Orkas のサーバーを経由してプロキシされることはありません。

Is Orkas a Claude Code or Cline alternative?Orkas é uma alternativa Claude Code ou Cline? Orkas 是 Claude Code 或 Cline 的替代品吗? Orkas は Claude Code や Cline の代替ですか?

Orkas is not only a terminal or editor coding agent. It is a local-first desktop workspace where native agents and external CLI coding agents can be coordinated from one conversation, so it complements or replaces single-purpose tools depending on whether you need a broader multi-agent workflow. See Orkas vs Claude Code and Orkas vs Cline.Orkas não é apenas um agente de codificação de terminal ou editor. É um espaço de trabalho desktop local-first onde agentes nativos e agentes de codificação CLI externos podem ser coordenados a partir de uma conversa, portanto, complementa ou substitui ferramentas de propósito único, dependendo se você precisa de um workflow multiagente mais amplo. Veja Orkas vs Claude Code e Orkas x Cline. Orkas 不只是终端或编辑器里的编程 agent。它是一个本地优先的桌面工作台,可以在同一段对话里协调原生 agent 和外部 CLI coding agent;如果你需要更宽的多 agent 工作流,它可以补充或替代单一用途工具。可阅读 Orkas vs Claude CodeOrkas vs Cline Orkas は単なるターミナル型、またはエディター内のコーディングエージェントではありません。ネイティブエージェントと外部 CLI コーディングエージェントを一つの会話から調整できるローカルファーストのデスクトップワークスペースです。より広いマルチエージェントワークフローが必要なら、単一目的のツールを補完または置き換えられます。Orkas vs Claude CodeOrkas vs Cline を参照してください。

Put a team on your repo

Free, open source, runs on your machine.