Basis

tech.basis1由 basis1.tech 发布官方发布者v0.2.2更新于 Nov 6, 2026

The Excel engine for AI agents: read, check and edit Excel models over MCP.

已验证Streamable HTTP可网页运行文件与存储开发者工具数据与分析

概览

AI 生成的概览

Basis 让助手通过 Basis API 解析、审计、重算和编辑 Excel 工作簿,并可用会话逐步修改输入、读取计算结果。

可以做什么
  • 把工作簿解析为 NDJSON 记录,提交异步解析任务、流式读取结果,并逐单元格比较两个工作簿
  • 用给定的单元格数值计算单条 Excel 公式
  • 重算 .xlsx 并保存公式结果,可选把结果写回本地文件
  • 以会话方式打开工作簿:设置输入、读取输出、定义名称、新增或重命名工作表、开关迭代计算
  • 检查模型结构,读取单元格或区域的公式与数值,追踪前置与后置依赖,审计错误与循环引用
  • 为会话创建检查点并回滚,导出或快照为 .xlsx,并保存到 Basis Library
适用场景
适合需要处理真实 Excel 模型的场景:读取或比较工作簿、确认公式算出什么、修复由 openpyxl 等库写出但没有结果的公式文件,或对财务与规划模型做假设调整并读取输出。
运行要求
需要 BASIS_API_KEY 中的 Basis API 密钥;使用托管端点时改用 Authorization 的 bearer 头。本地包需要 Node.js 20.19 或更高版本并通过 npx 运行,且需要网络访问。BASIS_BASE_URL 可指向自建服务器,BASIS_MCP_MAX_FETCH_BYTES 限制下载工作簿的大小。
安装前请注意
所有工具都在 Basis API 上执行:传入的工作簿、公式和单元格数值会上传到 api.basis1.tech,通过 path 传入的文件也会被读取并上传。会话在工作期间保留在服务器上,并会检查点保存到私有云存储;保存到 Library 的文件会一直保留直到删除。开启 overwrite 时重算和导出会覆盖本地文件。不能离开本机的工作簿不要使用此服务器。

安装

在 SourceWeft 中

  1. 打开 控制台中的 Basis,将其添加到工作区。
  2. 连接 Authorization token 凭据。它们会被私密存储,绝不会在公开页面上展示。
  3. 为需要使用其工具的对话启用该服务。

Web executable,通过 Streamable HTTP。 远程服务在工作区中配置后即可从网页运行时运行。

交给智能体安装

把这段提示词粘贴到 Claude Code、Codex、Cursor 等智能体,或 SourceWeft 对话中。智能体会读取该服务器的安装指南,先向你展示发布者、端点和登录方式,经你同意后再把它添加到你的客户端。密钥不会写进任何命令。

阅读 https://sourceweft.com/mcp/tech.basis1%2Fmcp/install.md,并按其中说明把这个 MCP 服务器接入我的客户端。修改任何配置之前,先告诉我它的发布者、端点和登录方式,等我确认。不要把 API 密钥写进命令或我会提交的文件;问我密钥应从哪里获取。

查看智能体所依据的安装指南

其他 MCP 客户端

把它添加到你客户端的 mcpServers 配置中。

{
  "mcpServers": {
    "basis": {
      "type": "http",
      "url": "https://api.basis1.tech/mcp",
      "headers": {
        "Authorization": "Bearer <YOUR_TOKEN>"
      }
    }
  }
}

README

@basis1/mcp

The Excel engine for AI agents. Let AI agents read, check and edit Excel models.

An MCP server for Basis. It lets Claude, Cursor and other MCP clients parse Excel workbooks, evaluate formulas, trace dependencies, and open a workbook as a session to change inputs and read results.

Every tool runs on the Basis API. You need a Basis API key and a network connection, and the workbooks you give a tool are uploaded to the Basis API at https://api.basis1.tech. See What gets uploaded.

Install

You need Node.js 20.19 or later and a Basis API key, which you create yourself in your account settings at basis1.tech/settings. Each snippet below starts the server with npx, which downloads the package the first time.

Claude Desktop

Open Settings → Developer → Edit Config and add this to claude_desktop_config.json, then restart Claude Desktop:

json
{  "mcpServers": {    "basis": {      "command": "npx",      "args": ["-y", "@basis1/mcp"],      "env": {        "BASIS_API_KEY": "your-api-key"      }    }  }}

Claude Code

sh
claude mcp add basis --env BASIS_API_KEY=your-api-key -- npx -y @basis1/mcp

Add --scope user to use it in every project, or --scope project to share it through the project's .mcp.json. In a shared file, write "BASIS_API_KEY": "${BASIS_API_KEY}" so each person's key comes from their own environment.

Cursor

Add this to ~/.cursor/mcp.json, or to .cursor/mcp.json in a project:

json
{  "mcpServers": {    "basis": {      "command": "npx",      "args": ["-y", "@basis1/mcp"],      "env": {        "BASIS_API_KEY": "your-api-key"      }    }  }}

Other MCP clients

Configure a stdio server with the command npx and the arguments -y @basis1/mcp, and set BASIS_API_KEY in its environment. On Windows, if your client can't start npx directly, use the command cmd with the arguments /c npx -y @basis1/mcp.

Without installing anything: the hosted endpoint

If your client supports remote MCP servers over HTTP with custom headers, you can connect to the hosted endpoint directly instead of running this package:

json
{  "mcpServers": {    "basis": {      "type": "http",      "url": "https://api.basis1.tech/mcp",      "headers": { "Authorization": "Bearer your-api-key" }    }  }}

In Claude Code: claude mcp add --transport http basis https://api.basis1.tech/mcp --header "Authorization: Bearer your-api-key".

It offers the same tools, except that it can't read files on your computer: pass a url or base64 instead of a path. This package adds path support by reading the file on your machine and uploading it for you.

Settings

VariableDefaultWhat it does
BASIS_API_KEYnone (required)Your Basis API key. Without it the server still starts, and every tool tells you the key is missing.
BASIS_BASE_URLhttps://api.basis1.techThe Basis API to use, for example a self-hosted server. It must use https:// unless it points at localhost, so your key is never sent unencrypted.
BASIS_MCP_MAX_FETCH_BYTES200000000The largest workbook a tool will download when you pass a url.

BASIS_TOKEN, the old name for the key, still works and logs a reminder to rename it.

If a setting is invalid, the server doesn't start: it writes what to fix to stderr, which your MCP client shows in its server logs (Claude Desktop: Settings → Developer, or ~/Library/Logs/Claude/mcp-server-basis.log on macOS). The server writes diagnostics only to stderr, because stdout carries the MCP protocol, and it never prints your key.

npx -y @basis1/mcp --help lists these settings, and --version prints the version.

Tools

Workbook inputs accept exactly one of path (a file on your computer), url or base64.

Parsing

  • parse_workbook: returns the workbook as NDJSON records (workbook_meta, sheet, cell, edge, image, image_summary, audit). Large files run as a background job and are polled to completion for you.
  • submit_job: starts a parse job and returns its handle (job_id, status_url, stream_url) without waiting.
  • get_job_status: the status of a job.
  • stream_job: the NDJSON output of a finished job, optionally from record since.
  • diff_workbooks: cell-level differences between two workbooks: added, removed and changed cells (value and/or formula), with format-only changes listed separately. Takes before_* and after_* inputs and include_format.

Formulas

  • evaluate_formula: evaluates one Excel formula against cell values you pass, for example {"formula": "NPV(0.08,A1:A5)", "data": {"A1": 100, ...}}. Supports 278 Excel functions, including financial, statistical, lookup and dynamic-array functions.

Recalculating a file

  • recalc_workbook: takes an .xlsx (path, url or base64) and returns it with computed values saved in every formula cell, plus the same error summary. Call it after you create or edit a workbook with openpyxl (or any library that saves formulas without results): without it, pandas, openpyxl data_only=True and other readers that are not Excel see empty cells. It replaces running LibreOffice's recalc.py. Pass output_path to write the result on your computer; the same path as the input with overwrite: true recalculates a file in place. The file is uploaded for the calculation and not kept.

Both file-returning tools carry over what Basis does not model (styles, charts, images, data validation, conditional formats, defined names, comments, VBA in .xlsm) byte for byte. What cannot be kept is listed under summary.warnings: per-word formatting inside a cell's text, what-if data tables, and the cached results of multi-cell array formulas. Formulas Basis cannot compute are listed under summary.uncomputed instead of being saved blank without notice.

Workbook sessions. A session holds one workbook open on the Basis API so you can change it step by step.

  • create_session: opens a workbook from path, url, base64 or your Library (library_file_id), or starts a blank one. Optional calc turns on iterative calculation. A local path travels inside one request, so files over about 24 MB can't open as a session yet; parse_workbook handles large files.
  • list_sessions, close_session.
  • set_inputs: sets values or formulas by defined name or cell reference, recalculates, and reports which named outputs changed.
  • get_outputs: reads computed values by name, cell reference or range (Sheet1!H2:H18), or every defined name.
  • define_variable: creates or retargets a defined name.
  • add_sheet: adds a sheet to a blank workbook.
  • rename_sheet: renames a sheet; formulas, defined names and validations that point at it follow the new name.
  • set_calc_settings: turns iterative calculation on or off for circular models.
  • inspect_model: sheets, defined names, calculation settings and statistics, optionally with the dependency graph.
  • get_cell_formula, get_formula_range: formulas and values for one cell or a range.
  • trace_dependencies: precedents and dependents of a cell, up to 5 levels deep.
  • audit_workbook: a structural audit (errors, broken references, cycles, named ranges).
  • commit_session, restore_session: named checkpoints and rollback.
  • export_workbook: recalculates the session and returns it as an .xlsx with every formula's computed value saved in the file, plus a summary of formula errors (#REF!, #DIV/0!, #NAME? with sheet and cell). Pass output_path to write the file on your computer instead of getting base64 back (overwrite: true to replace an existing file).
  • snapshot_session: the current workbook as .xlsx bytes (base64), no summary. Prefer export_workbook.
  • save_to_library: saves the workbook to your Basis Library as a new file or a new version.

Tool errors come back as a tool result with isError set and a JSON body {"code", "message", "status"}, for example missing_api_key, unauthorized, network_error, session_not_found or one of the API's error codes.

What gets uploaded

  • Parsing and diffing: the workbook is sent to the Basis API. Parse jobs and their results are kept in the server's memory only and expire after 15 minutes.
  • Formulas: the formula and the cell values you pass are sent to the Basis API.
  • Sessions: the workbook is uploaded when you create the session and stays on the Basis API while you work. A session closes after 30 minutes without use, or when you call close_session. So that sessions survive server restarts, their state is also checkpointed to private cloud storage, which deletes checkpoints within a day.
  • Library: save_to_library stores the workbook in your Library until you delete it.
  • Files on your computer: a tool reads a path you pass and uploads that file's contents. It reads nothing else.
  • URLs: this package downloads a url you pass to parse_workbook, submit_job or diff_workbooks on your machine and uploads the bytes. For sessions, the Basis API downloads the URL itself.

All traffic uses HTTPS, with your API key in the Authorization header. If your workbooks can't leave your machine, don't use this server with them.

Development

From a checkout of this repository:

sh
(cd ../sdk && npm ci && npm run build)  # @basis1/parse, linked as file:../sdknpm cinpm run build                            # typecheck, then bundle dist/index.jsnpm test                                 # unit, hosted forwarding and stdio handshake testsnpm run test:pack                        # pack, install in a temp dir, MCP handshake over stdio

To run your build against another Basis server, for example a self-hosted one:

sh
BASIS_API_KEY=devtoken BASIS_BASE_URL=http://localhost:8080 node dist/index.js

How the package is built

dist/index.js is a single esbuild bundle that includes @modelcontextprotocol/sdk and zod (both MIT). As runtime dependencies they would install 95 packages (23 MB) on every first npx run, including web-server frameworks a stdio server never uses. Bundled, npx installs two packages and runs the exact versions we tested. Their license texts ship in dist/THIRD_PARTY_LICENSES.txt, and the build fails if a bundled package isn't permissively licensed.

@basis1/parse is a real runtime dependency, not bundled. In the repository it's file:../sdk. The release process rewrites that to ^<sdk version> before publishing, and scripts/check-publishable.mjs refuses a package that still has a local path.

Releasing

Releases are published by the maintainers. When a change touches both packages, the SDK is released first, then the MCP server, which depends on the published SDK version. version is kept in sync between package.json and server.json.

MCP Registry

server.json describes this server for the MCP Registry as tech.basis1/mcp, and mcpName in package.json links the npm package to it. The tech.basis1/ namespace belongs to the basis1.tech domain rather than a personal GitHub account, so the registry verifies it through DNS. Nothing is submitted automatically.

The maintainers list each release in the registry after the npm publish, signing in with mcp-publisher login dns --domain basis1.tech. The tech.basis1/ namespace is verified through a DNS record on basis1.tech.

License

Apache-2.0. See LICENSE.

来源:mcp/README.md,提交 7e0f9c8

工具

26

parse_workbook

Parse workbook

Parse an .xlsx workbook and return the full semantic NDJSON output (workbook_meta, sheet, cell, edge, image, image_summary, audit records).

输入参数urlbase64webhook_url

submit_job

Submit parse job

Submit an .xlsx workbook for async parsing and get back a job handle (job_id, status_url, stream_url) without waiting for it to finish.

输入参数urlbase64webhook_url

get_job_status

Get job status

Check the status of a parse job submitted via submit_job.

输入参数job_id必填

stream_job

Stream job output

Fetch the NDJSON output of a completed parse job. Pass `since` to resume from a record index.

输入参数sincejob_id必填

diff_workbooks

Diff workbooks

Compare two .xlsx workbooks and return the cell-level differences as JSON. Cells are joined on (sheet title, cell ref) and classified into added, removed, and changed (value and/or formula), with format-only changes bucketed separately. Provide exactly one of before_url/before_base64 and one of after_url/after_base64.

输入参数after_urlbefore_urlafter_base64before_base64include_format

evaluate_formula

Evaluate formula

Evaluate an Excel formula locally and return the computed result: a quick way to check what a formula returns on a few sample values before you put it in a workbook. Pass data as a map of cell references to values (e.g. {"A1": 100, "A2": 200, "B1": "hello"}) and a formula string (e.g. "SUM(A1:A2)"). The formula is placed in row 1 of the first column right of your data, so don't reference that cell. A formula that evaluates to an Excel error returns type "error" with the code (#VALUE!, #N/A…) as a normal result; a formula that can't be parsed or uses an unsupported function is a tool error. Supports 278 Excel functions: math, financial (NPV, IRR, XIRR, PMT…), statistical, lookup, date/time, text, logical, and dynamic arrays (FILTER, SORT, UNIQUE, SEQUENCE…). Array expressions evaluate element by element inside SUMPRODUCT, LOOKUP, AGGREGATE, SMALL/LARGE, MATCH and INDEX. To check formulas that already live in a workbook, use create_session instead.

输入参数dataformula必填

create_session

Create session

Create a persistent computation session for a workbook. Pass library_file_id to load from the Library, base64 for raw xlsx bytes, url to fetch from a URL, or omit all for a blank model. Returns a session_id for subsequent calls. The workbook is recalculated on load, including formulas written by other tools that carry no cached values, so open a file you have edited to read back what its formulas compute (get_outputs, get_formula_range), or change inputs with set_inputs.

输入参数urlcalcbase64document_idlibrary_file_id

close_session

Close session

Close a computation session and release its resources.

输入参数session_id必填

list_sessions

List sessions

List all active computation sessions for the current user.

无需输入参数。

add_sheet

Add sheet

Add a new sheet to a session's workbook. The new sheet becomes the active sheet. Only works on sessions created from blank documents (not imported xlsx files).

输入参数name必填session_id必填

rename_sheet

Rename sheet

Rename a sheet in a session's workbook. Formulas, named ranges and validations that reference the sheet are rewritten to the new name, so results don't change. Excel's rules apply: 1-31 characters, none of : \ / ? * [ ], no leading or trailing apostrophe, unique ignoring case.

输入参数name必填session_id必填sheet_index必填

set_inputs

Set inputs

Set one or more cells in a session and recalculate. Keys can be defined names (e.g. "Revenue") or cell references (e.g. "Sheet1!A1"). A string starting with "=" is stored as a formula, so you can also try a formula in place and read the result with get_outputs. Returns the applied changes and which named outputs changed (effects).

输入参数inputs必填session_id必填

set_calc_settings

Set calc settings

Turn iterative calculation on or off for a session and recalculate. Needed for circular models (e.g. development pro formas where loan size depends on cost, which includes the loan's interest and fees): with it on, set_inputs re-converges the loop on every call. Saved with the session's next commit (restore_session to an earlier commit rolls it back) and written to the workbook, so an exported xlsx iterates in Excel too.

输入参数iterative必填max_changesession_id必填max_iterations

get_outputs

Get outputs

Read computed values from a session, e.g. to check what the formulas in a workbook return. Pass defined names, cell refs ("B4", "Sheet1!B4") or ranges ("Sheet1!H2:H18", one entry per cell), or omit to read all defined names. A cell that evaluates to an Excel error is returned with type "error", not as a tool failure.

输入参数namessession_id必填

define_variable

Define variable

Create or retarget a named variable in the session's workbook. The name can then be used in set_inputs and get_outputs instead of cell references.

输入参数name必填refers_to必填session_id必填scope_sheet

inspect_model

Inspect model

Inspect a session's model structure: sheets, defined names, cell/formula counts, and optionally the full dependency edge graph.

输入参数session_id必填include_edges

snapshot_session

Snapshot session

Export the session's current state as xlsx bytes (base64). Optionally creates a commit first. To get the file for a user or a non-Excel reader, with formula errors listed, prefer export_workbook.

输入参数session_id必填commit_message

export_workbook

Export workbook

Get a session's workbook back as an .xlsx file with every formula's computed value saved in it. Recalculates first. Call this when you need the file itself: to give it to the user, or to read it with pandas, openpyxl (data_only=True) or any tool that is not Excel. Returns file_base64 and a summary of formula errors (#REF!, #DIV/0!, #NAME? with sheet and cell) and of anything that did not survive the save. Everything Basis does not model (styles, charts, images, data validation, defined names, comments, VBA) is carried over from the original file unchanged.

输入参数session_id必填

recalc_workbook

Recalculate workbook

Recalculate an .xlsx file and return it with the computed values saved in every formula cell. Use this after you create or edit a workbook with openpyxl or any other library that writes formulas without results: those files have no cached values, so pandas, openpyxl (data_only=True) and other readers see empty cells until the file is recalculated. Replaces running LibreOffice recalc.py. Returns file_base64 plus a summary of formula errors (#REF!, #DIV/0!, #NAME?, #VALUE!... with sheet and cell) and of formulas Basis could not compute: fix those and call again. Styles, charts, images, data validation, defined names, comments and VBA are carried over unchanged. Pass exactly one of url or base64.

输入参数urlbase64filename

get_formula_range

Get formula range

Batch-read formulas and computed values for a rectangular range of cells: the way to check what the formulas in a file compute. Returns {cell, formula, value, value_type} per cell; a formula that errors has value_type "error". Much more efficient than calling get_cell_formula repeatedly.

输入参数range必填session_id必填

commit_session

Commit session

Create a named checkpoint of the session's current state. Returns the commit ID which can be passed to restore_session to roll back. Checkpoints live in the session only; to keep one in the Library's version history, call save_to_library after it (the version takes this message as its name).

输入参数message必填session_id必填

restore_session

Restore session

Roll back a session to a previous commit. Reverts all cell changes made since that commit. Returns the number of cells that were changed by the rollback.

输入参数commit_id必填session_id必填

get_cell_formula

Get cell formula

Return the formula text and computed value for a single cell. If the cell has no formula (constant or empty), formula will be null.

输入参数cell必填session_id必填

trace_dependencies

Trace dependencies

Trace the dependency graph from a cell. direction="precedents" shows what feeds into this cell; direction="dependents" shows what this cell feeds. Returns a BFS traversal up to the given depth with formula text and computed value at each node.

输入参数cell必填depthdirection必填session_id必填

audit_workbook

Audit workbook

Run a structural audit on the session's workbook. Returns formula errors, broken references, circular dependencies, hardcoded plugs, and verification mismatches.

输入参数session_id必填

save_to_library

Save to Library

Save the session's workbook to the user's Library for durable storage. If the session was loaded from the Library or was previously saved, this creates a new version of that file. Otherwise it creates a new Library file. Returns the library_file_id. Each save is a named version in the Library's history: pass `message`, or it takes the latest commit_session message since the last save. commit_session checkpoints themselves stay in the session, so save after each step you want kept as a version.

输入参数namemessagesession_id必填

版本历史

3
  1. v0.2.2最新Nov 6, 2026
  2. v0.2.1Nov 5, 2026
  3. v0.1.0Nov 4, 2026

评分

已满足适用检查项的 100%

该评分衡量此收录条目的完整程度,依据注册中心条目、端点与公开来源计算得出。它不是对服务器本身的评价,也不表示服务器安全或当前可用:可连通性是带日期的检查结果,评分也不包含安装量或用户评价。

  • 近 30 天内可连通必需25 分
  • 已声明认证方式必需10 分
  • 已列出工具必需15 分
  • 描述必需15 分
  • 图标10 分
  • 文档15 分
  • 许可证10 分