把 JetBrains Air 連接到 Busabase
JetBrains' system for building software with agents. Air in IDEs runs coding agents such as Claude Agent, Codex, Gemini CLI, Junie and GitHub Copilot — plus any agent connected via the Agent Client Protocol (ACP) — side by side, with every change reviewable in context. Air Teams runs shared projects and automations in dedicated cloud environments, and Air Governance controls which agents, models and MCP servers an organization uses.
把這段貼進 JetBrains Air
請閱讀 https://busabase.com/SETUP_SKILL.md,並幫我設定好 Busabase。請照著文件一路做到完成,最後確認連線是否正常運作。

把這一句指令交給你的 Agent
你得到的不只是一個連接器:busabase 技能還帶著完整的變更請求工作流程、CLI 與完整 API —— 之後每一次工作階段都在,不必再重新貼一次。
會先問你
Busabase Cloud 還是個人桌面版 —— 在動手之前,由你決定。
完成連線
幫你登入並自行儲存憑證,不需要手動複製 API 金鑰。
建立一個資料表
建立並填入你的第一個資料表,讓工作區一開始就不是空白畫面。
安裝技能
busabase 與 busabase-app-creator 會永久安裝,之後每一次工作階段都能直接使用。
此 Agent 的注意事項
Skills belong to the agent Air is running, not to Air itself: install the Busabase skill separately for each agent you use in Air (Claude Agent, Codex, Gemini CLI, Junie, Copilot). Air automations can also carry startup scripts, MCP servers and skills for the run, but the skill is still loaded by the underlying agent.
比較想用連接器?
如果你不想讓 Agent 自行安裝技能,而是想把 Busabase 當作 MCP 工具接進來,可以使用下面這個伺服器網址。
一覽表
- 廠商
- JetBrains
- 連接方式
- MCP + API 金鑰 · 外掛/技能
- 傳輸方式
- Streamable HTTP
- 驗證方式
- API 金鑰
- 設定檔
- .mcp.json (repository root, cloud tasks and automations)
Air runs other coding agents, and each of those agents can execute shell commands and fetch URLs, so the Skill path applies to the agent itself: start the agent in Air and ask it to Read https://busabase.com/SETUP_SKILL.md and set up Busabase for me. The agent installs the Busabase skill in its own usual location, and later Air sessions with that agent can use it.
Give every cloud task and automation the Busabase connector
For Air Teams cloud tasks and automations, add Busabase as an MCP server in a .mcp.json file at the repository root:
{
"mcpServers": {
"busabase": {
"url": "https://busabase.com/api/mcp",
"headers": {
"Authorization": "Bearer ${BUSABASE_API_KEY}"
}
}
}
}
Add the API key's value as a secret named BUSABASE_API_KEY in the environment configuration (names are case-sensitive) — never write the token into the file itself. Commit and push; the next cloud task or automation run on that branch gets the server. The Air app also configures MCP servers through its own settings (see the vendor's MCP page linked below).
JetBrains Air MCP documentation
已知的注意事項
Busabase is not on Air Teams' managed connector list (Confluence, Figma, GitHub, GitLab, Grafana, Jira, Linear, Sentry, TeamCity, YouTrack), so the `.mcp.json` route above is the way in; the connector list's sign-in flow does not apply to custom servers, and the vendor docs describe header-token auth for `.mcp.json` servers rather than OAuth, so auth-mode is api-key. If a managed connector and a `.mcp.json` entry share a name, the agent gets the connector and the `.mcp.json` entry is skipped. Project-level connectors do not reach cloud tasks — cloud tasks only use personal connectors.
能力比較
與目錄中其他已收錄的 Agent,並列比較連線能力。
| 能力 | |||
|---|---|---|---|
| 類型 | AI 編輯器 | 終端機 Agent | AI 編輯器 |
| 推薦路徑 | Agent Skill | Agent Skill | Agent Skill |
| 連接方式 | MCP + API 金鑰 · 外掛/技能 | MCP + OAuth · 外掛/技能 | MCP + OAuth · 外掛/技能 |
| 傳輸方式 | Streamable HTTP | Streamable HTTP | Streamable HTTP |
| 驗證方式 | API 金鑰 | OAuth(動態註冊) | OAuth(動態註冊) |
把這段貼進 JetBrains Air
請閱讀 https://busabase.com/SETUP_SKILL.md,並幫我設定好 Busabase。請照著文件一路做到完成,最後確認連線是否正常運作。