Pare Bun

Community

Structured JSON output for Bun runtime operations

How to install ↓ Compare GitHub source repository ↗

Category: Dev Tools #426 of 438
Stars: ★ 138 Popular
Transport: stdio (local process)
Runtime: Node.js 20+ · Bun
Credentials: No credential needed
License: MIT
Last commit: 53 days ago

26FMRS · D

Pare Bun is a package within the Pare collection, focused on turning Bun operations into structured JSON to improve AI agent efficiency. It is third-party, not an official Bun project. Users should be aware of command side effects and version compatibility.

Strongest · Setup experience 6/20 Weakest · Reliability 4/20

Reliability

4/20

Security and permissions

5/20

Maintenance

6/20

Documentation

5/20

Setup experience

6/20

Why each score

Reliability 4/20: Only the manifest and README were supplied; there is no source, test, or CI evidence for packages/server-bun to verify startup handshake, tool-list, or declared behavior. The manifest does not list tools or outputSchema. No contradiction is visible, so the score is low rather than zero.
Security and permissions 5/20: No red-line issue such as malware, credential theft, or irreversible actions is evident, but the Bun server executes runtime commands and there is no documentation of permission boundaries, confirmation mechanisms, or data flow, so least privilege cannot be established.
Maintenance 6/20: The repository is not archived, has an MIT license, and version 0.22.0 indicates releases, but there are no commit/release dates, security-response channel, changelog, or dependency-update evidence; activity from sibling servers cannot be attributed to this one.
Documentation 5/20: The README provides general architecture and setup guidance, but no server-bun-specific tool parameters, limits, troubleshooting, or response examples; per evidence rules, other servers' documentation cannot be counted.
Setup experience 6/20: The npm package name, stdio transport, and Node>=20 requirement are present, but there is no clear client configuration example for the bun server, no platform compatibility evidence, and no CI/tests to corroborate a working install; the static cap of 15 also limits the score.

Static review · not runListed 2026-08-07

Read the FMRS scoring method →

Fit and risk

What it can accessRuns commands or code

Best for

Projects using Bun as the JavaScript runtime and package manager
Teams relying on AI agents for coding, building, and testing automation
Developers focused on token efficiency and output reliability

Not for

Scenarios that do not require Bun or structured output
Use cases needing other runtime tools
Environments that rely on interactive terminal output

Required permissions

Permissions to execute Bun commands
Read and write access to project files (depending on usage)
Network access for downloading packages or interacting with remote repositories as needed

Risks and side effects

Executing Bun commands may have side effects such as modifying files or installing packages
Output schemas could change with version updates, potentially breaking existing agent logic
Not maintained by Bun's upstream project; compatibility issues may arise

Setup

Before you start

Runtime：Node.js 20+ · Bun

Other optional settings (1)

PARE_BUN_TOOLS optional Optional per-server tool filter listing Bun tool names to register, comma-separated; empty disables all tools, unset enables all.

Configure your MCP client to launch the server with the command 'npx -y @paretools/bun'
Optionally restrict tools via the environment variable PARE_BUN_TOOLS
Restart your client session.

claude_desktop_config.json

{
  "mcpServers": {
    "pare-bun": {
      "command": "npx",
      "args": [
        "-y",
        "@paretools/bun"
      ],
      "env": {
        "PARE_BUN_TOOLS": ""
      }
    }
  }
}

Shown for Claude Desktop. Other clients may use a different file or key (VS Code uses "servers") — the configurator below converts it.

.vscode/mcp.json

{
  "servers": {
    "pare-bun": {
      "command": "npx",
      "args": [
        "-y",
        "@paretools/bun"
      ],
      "env": {
        "PARE_BUN_TOOLS": ""
      }
    }
  }
}

Goes in your project's .vscode/mcp.json (VS Code uses a "servers" key).

Terminal

claude mcp add pare-bun -e PARE_BUN_TOOLS= -- npx -y @paretools/bun

Run it in a terminal; replace any <…> placeholders with your own values first.

Generate and validate client config

Check that it works

After starting the client, confirm the pare-bun server (@paretools/bun) tools appear in the client's tool list. Note the sample config sets PARE_BUN_TOOLS to an empty string, which disables all tools — set it to tool names like "install,run,test" to see them.

Troubleshooting

Ensure Node.js version >= 20
If npx fails, try installing '@paretools/bun' globally or use an absolute path
If connection timeouts occur, increase MCP_TIMEOUT in your client config
Check whether environment variable PARE_BUN_TOOLS limits tools, potentially making the server unusable

Things to try

Once connected, you can ask your AI assistant things like:

Run this project's tests with Bun and give me a structured summary of the results
List the Bun packages installed in this project
Run the test script with bun and tell me which tests failed
Run bun install and report whether it succeeded

Use cases

AI agents that need to run Bun operations (e.g., installing packages, running scripts) and consume structured output directly

Automated development workflows aiming to reduce token consumption

Development environments requiring consistent, cross-platform command output

Supported clients

Claude Desktop

Claude Code

Cursor

VS Code

Windsurf

Cline / Roo Code

Codex

Gemini CLI

Zed

Continue

Listed from the project's documentation, not tested by this site.

Overview

Pare Bun is an MCP server from the Pare project that wraps Bun runtime operations into typed, schema-validated JSON, replacing fragile terminal text parsing. Using MCP's structuredContent and outputSchema features, it provides reliable, directly actionable data for AI agents, reducing token usage and enhancing cross-platform compatibility.

Similar servers

Context7 80 · B

Upstash's official server providing up-to-date third-party library docs for AI coding assistants

★ 62.9k · Tools 2 Compare with this →

NPM Sentinel MCP Server 79 · B

AI-powered NPM package analysis MCP server

★ 18 · Tools 19 Compare with this →

MartinLoop MCP Server 79 · B

An independent control layer that gives AI coding agents budgets, brakes, and receipts

★ 410 Compare with this →

GitHub MCP Server 76 · B

GitHub's official MCP server for managing repos, issues, PRs, and workflows via natural language

★ 33.5k · Tools 15 Compare with this →

Compare with other MCP servers

More servers from this repository

Source revision 57e1980e8375 Data synced 2026-10-12 Read the FMRS scoring method