Skip the install. Get this working in under 2 minutes.
Start a free trial on cloud.anythingmcp.com, add the Coda in one click, then point your AI client (Claude, ChatGPT, Copilot or Cursor) at the generated MCP endpoint. No Docker, no git clone, zero engineering experience required.
Summary
Drive Coda (docs + tables + formulas) from any AI agent: docs, tables, rows, columns, formulas, packs, controls. 13 tools, Bearer token auth.
试试这样问
Coda 的示例提示
点击任意提示即可复制，然后粘贴到 Claude、ChatGPT、Cursor、Gemini、Copilot 或 OpenClaw 中，在该连接器上运行。
Claude 是 AI，可能会出错。请核对回答。
💡 无需安装? 直接使用 cloud.anythingmcp.com。 登录、点击 Connectors → Coda、粘贴凭证、生成 MCP API key — 完成。无 Docker、无
git clone、无本地服务器。
Coda + ChatGPT
Drive Coda (docs + tables + formulas) from any AI agent: docs, tables, rows, columns, formulas, packs, controls. 13 tools, Bearer token auth.
前提条件
完整的设置说明已内置在连接器中 (在 store 中选择连接器时可见)。所需环境变量:
CODA_API_TOKEN
步骤 1 — 获取凭证
io/developers/apis/v1).
Setup:
- Sign in to https://coda.io → top-right avatar → Account → API settings → Generate API token.
- Name the token. Copy it. Set
CODA_API_TOKEN.
Authentication: Authorization: Bearer ${CODA_API_TOKEN}.
Resource hierarchy: Doc → Section (Page) → Table → Row → Cell. Coda's twist: tables are first-class with structured columns and types; rows are addressable individually; columns can hold formulas.
Doc IDs look like dXXXXXXXXXX (10-char prefix d). Visible in the doc URL.
Table IDs look like grid-XXXXX. Discover via coda_list_tables(docId).
…(continued in the in-app connector instructions)
步骤 2 — 安装 adapter
git clone https://github.com/HelpCode-ai/anythingmcp.git
cd anythingmcp && docker compose up -d
步骤 3 — 在 ChatGPT 中添加连接器
ChatGPT 通过 Connectors 功能 (Plus / Pro / Team / Enterprise 套餐) 支持 MCP 服务器。免费套餐不支持自定义连接器。
- 打开 ChatGPT → Settings → Connectors (或点击 composer 中的 🔧 图标 → Manage connectors)。
- 点击 "Add custom connector"。
- 填入:
- Name:
Coda - MCP server URL:
https://cloud.anythingmcp.com/mcp - Authentication: Bearer token → 粘贴 MCP API key
- Name:
- 点击 Connect。
可用工具
| Tool | What it does |
|---|---|
coda_whoami | Return the user the token belongs to |
coda_list_docs | List docs the user can access |
coda_get_doc | Fetch a single doc by ID |
coda_list_tables | List tables in a doc |
coda_list_columns | List columns in a table |
coda_list_rows | List rows in a table or view |
coda_get_row | Fetch a single row by ID |
coda_insert_rows | Insert (or upsert with keyColumns) rows into a table |
coda_update_row | Update a specific row by ID |
coda_delete_row | Delete a single row by ID |
coda_delete_rows | Bulk-delete rows by IDs (more efficient than per-row) |
coda_get_formula_result | Evaluate a named formula on a doc |
coda_get_mutation_status | Check the status of an async mutation (insert/update/delete) by its requestId returned from those calls |
FAQ
ChatGPT 免费套餐能用自定义 MCP Connectors 吗? 不能 — 需要 Plus、Pro、Team 或 Enterprise。
下一步
这份指南对你有帮助吗?
