PluginBench
MCP Server
Active
MIT

io.github.Dave-London/pare-http MCP Server

io.github.Dave-London/pare-http

Structured HTTP requests (GET, POST, HEAD) as typed JSON for AI agents — no parsing fragile responses.

What is the io.github.Dave-London/pare-http MCP server?

The Pare HTTP MCP server wraps HTTP request operations (GET, POST, HEAD) and returns clean, schema-validated JSON instead of raw response text. It eliminates brittle string parsing by providing typed, consistent data that agents can act on directly.

Pare HTTP provides reliable, structured HTTP request capabilities for AI agents. Instead of raw HTTP responses with headers, encoding issues, and formatting variations, it returns validated JSON with consistent field names. This reduces token usage and eliminates parsing errors across different servers and response formats.

How to install io.github.Dave-London/pare-http

Copy-paste configuration for popular MCP clients.

transport: stdio
Config generated by PluginBench — verify against the source before use.
~/Library/Application Support/Claude/claude_desktop_config.json
{
  "mcpServers": {
    "pare-http": {
      "command": "npx",
      "args": [
        "-y",
        "@paretools/http"
      ]
    }
  }
}

Tools & capabilities

Tools this server exposes to the agent.

  • GET — Perform HTTP GET requests and return structured response data as typed JSON
  • POST — Perform HTTP POST requests with payload support and return structured response data as typed JSON
  • HEAD — Perform HTTP HEAD requests to retrieve headers without response body and return structured metadata as typed JSON

Use cases

  • Fetch API data and parse responses reliably without string manipulation
  • Submit form data or JSON payloads via POST with validated structured responses
  • Check endpoint availability and retrieve headers via HEAD requests
  • Integrate external APIs into agent workflows with type-safe, schema-validated data
  • Reduce token usage by receiving compact JSON instead of raw HTTP responses

io.github.Dave-London/pare-http MCP server FAQ

What is the Pare HTTP MCP server?

Pare HTTP wraps HTTP request operations (GET, POST, HEAD) and returns schema-validated JSON instead of raw response text. It eliminates brittle parsing and provides typed data agents can use directly.

Is Pare HTTP free?

Yes, Pare is open-source under the MIT license and free to use.

How do I install Pare HTTP in Cursor or Claude?

Install via npm: `npm install @paretools/http`. Then configure it in your MCP settings using the setup guide at https://github.com/Dave-London/Pare/docs/setup/ for your specific client (Cursor, Claude Desktop, Claude Code, etc.).

What authentication does Pare HTTP require?

Pare HTTP does not require authentication to the MCP server itself. Authentication depends on the HTTP endpoints you call — pass credentials via headers or request body as needed.

Can I filter which HTTP tools are available?

Yes, use environment variables like `PARE_HTTP_TOOLS` to restrict tools, or `PARE_TOOLS=http:GET,http:POST` for universal filtering across all Pare servers.

What are the token savings?

Pare returns compact, structured JSON instead of raw HTTP responses with headers, formatting, and decorative text. Typical savings range from 36–95% depending on response verbosity.

README (reference)

Source of truth, from the repository.

<h1><img src="assets/logo.png" alt="" width="80" valign="middle" />&nbsp;&nbsp;Pare</h1>

CI codecov npm Downloads TypeScript License: MIT Node.js >= 20 OpenSSF Scorecard OpenSSF Best Practices

Reliable, structured CLI output for AI agents — no more parsing fragile terminal text.

Pare provides MCP servers that wrap common developer tools (git, npm, docker, test runners, etc.) and return clean, schema-validated JSON instead of raw terminal text. Agents get typed data they can act on directly, without brittle string parsing.

The Problem

Parsing CLI output is fragile. Raw terminal text includes ANSI escape codes, decorative headers, progress bars, locale-specific formatting, and platform differences that break agent workflows in subtle ways. An agent that works fine with git status on macOS may fail on Windows because the output format changed. A test runner's summary line might shift between versions, silently breaking a regex.

Pare eliminates this entire class of errors by returning schema-validated JSON with consistent field names, regardless of platform, tool version, or locale. As a bonus, structured output is significantly smaller — agents use fewer tokens per tool call:

Tool CommandRaw TokensPare TokensReduction
docker build (multi-stage, 11 steps)3732095%
git log --stat (5 commits, verbose)4,99238292%
npm install (487 packages, warnings)2414183%
vitest run (28 tests, all pass)1963980%
cargo build (2 errors, help text)43613868%
pip install (9 packages, progress bars)28810165%
cargo test (12 tests, 2 failures)35119046%
npm audit (4 vulnerabilities)28718536%

Token estimates use ~4 chars/token. The biggest savings appear on verbose commands (builds, installs, tests). For simpler tools like eslint or tsc, the main advantage is reliable structured data — agents can use typed JSON directly rather than parsing strings.

How It Works

Each Pare tool returns two outputs:

  • content — human-readable text, for MCP clients that display it
  • structuredContent — typed, schema-validated JSON, ready for agents to process

This uses MCP's structuredContent and outputSchema features to provide type-safe, validated data that agents can rely on without custom parsing.

Example: git status

Raw git output (~118 tokens):

On branch main
Your branch is ahead of 'origin/main' by 2 commits.
  (use "git push" to publish your local commits)

Changes to be committed:
  (use "git restore --staged <file>..." to unstage)
        modified:   src/index.ts
        new file:   src/utils.ts

Changes not staged for commit:
  (use "git add <file>..." to update what will be committed)
  (use "git restore <file>..." to discard changes in working directory)
        modified:   README.md

Untracked files:
  (use "git add <file>..." to include in what will be committed)
        temp.log

Pare structured output (~59 tokens):

{
  "branch": "main",
  "upstream": "origin/main",
  "ahead": 2,
  "staged": [
    { "file": "src/index.ts", "status": "modified" },
    { "file": "src/utils.ts", "status": "added" }
  ],
  "modified": ["README.md"],
  "deleted": [],
  "untracked": ["temp.log"],
  "conflicts": [],
  "clean": false
}

50% fewer tokens. Zero information lost. Fully typed. Savings scale with output verbosity — test runners and build logs see 80–92% reduction.

Available Servers (28 packages, 240 tools)

Install only the servers relevant to your stack — most projects need just 2–4. The full catalog covers a wide range of ecosystems so Pare works wherever you do.

CategoryServersToolsWraps
Version Controlgit, github55git, gh
Languages & Packagesnpm, python, cargo, go, deno, bun, nix, dotnet, ruby, swift, jvm101npm, pip, cargo, go, deno, bun, nix, dotnet, gem, swift, gradle, maven
Build, Lint & Testbuild, lint, test, cmake, bazel23tsc, esbuild, vite, webpack, eslint, prettier, biome, vitest, pytest, jest
Infrastructuredocker, k8s, infra, security, remote40docker, kubectl, helm, terraform, ansible, trivy, ssh
Utilitiessearch, http, make, process, db21ripgrep, fd, curl, make, just, psql, mysql, redis, mongosh

Tool Schemas — detailed response examples and field descriptions for every tool. See also Tool Response Examples for quick JSON samples.

Quick Setup

# 1. Configure MCP servers (non-interactive)
npx @paretools/init --client claude-code --preset web

# 2. Add agent rules to your project
#    (append to existing CLAUDE.md, or copy if new)
cat node_modules/@paretools/init/rules/CLAUDE.md >> CLAUDE.md

# 3. Restart your client session

# 4. Validate
npx @paretools/init doctor

Available presets: web, python, rust, go, jvm, dotnet, ruby, swift, mobile, devops, full

Setup Guides by Client

Claude CodeClaude DesktopCursor
VS Code / CopilotWindsurfCline / Roo Code
OpenAI CodexGemini CLIZed
Continue.dev

Full Quickstart Guide — presets, ecosystem mapping, validation

Manual Configuration — config paths and formats for all clients

Agent Integration Guide — rule files, hooks, CLI-to-MCP mapping

Configuration

Tool Selection

By default, every Pare server registers all of its tools. If a server exposes tools you don't need — or you want to limit which tools are available to an agent — you can filter them with environment variables.

Per-server filter — restrict a single server's tools:

# Only register status and log in the git server
PARE_GIT_TOOLS=status,log npx @paretools/git

Universal filter — restrict tools across all servers:

# Only register these specific tools across any server
PARE_TOOLS=git:status,git:log,npm:install npx @paretools/git

Disable all tools — set the env var to an empty string:

PARE_GIT_TOOLS= npx @paretools/git   # no tools registered
Env VarScopeFormatExample
PARE_TOOLSAll serversserver:tool,...git:status,npm:install
PARE_{SERVER}_TOOLSOne servertool,...status,log,diff

Rules:

  • No env var = all tools enabled (default)
  • PARE_TOOLS (universal) takes precedence over per-server vars
  • Server names use uppercase with hyphens replaced by underscores (e.g., PARE_MY_SERVER_TOOLS)
  • Whitespace around commas is ignored

Common patterns:

# Read-only git (no push, commit, add, checkout)
PARE_GIT_TOOLS=status,log,diff,branch,show

# Minimal npm
PARE_NPM_TOOLS=install,test,run

# Only specific tools across all servers
PARE_TOOLS=git:status,git:diff,npm:install,test:run

In JSON MCP config, set via the env key:

{
  "mcpServers": {
    "pare-git": {
      "command": "npx",
      "args": ["-y", "@paretools/git"],
      "env": {
        "PARE_GIT_TOOLS": "status,log,diff,show"
      }
    }
  }
}

Troubleshooting

IssueSolution
npx not found / ENOENT on WindowsUse cmd /c npx wrapper (see your client's setup guide)
Slow first startRun npx -y @paretools/git once to cache, or install globally: npm i -g @paretools/git
Node.js version errorPare requires Node.js >= 20
NVM/fnm PATH issuesUse absolute path to npx: e.g., ~/.nvm/versions/node/v22/bin/npx
MCP connection timeoutSet MCP_TIMEOUT=30000 for Claude Code, or increase initTimeout in client config
Too many tools filling contextUse tool selection env vars to limit tools, or only install the servers you need

Contributing

Each server is a self-contained package. See CONTRIBUTING.md for the full guide.

License

MIT

Related MCP servers

Structured linting and formatting output as typed JSON for ESLint, Prettier, Biome, and Oxlint.

130
TypeScript
MIT
View repository →

Structured Make/Just task runner operations as typed JSON for AI agents.

130
TypeScript
MIT
View repository →

Structured npm/pnpm operations (install, audit, outdated, list, run) as typed JSON for AI agents.

130
TypeScript
MIT
View repository →

Structured Python tool output (ruff, mypy, pip, uv, black, pytest) as typed JSON for AI agents.

130
TypeScript
MIT
View repository →

Structured code search with ripgrep and fd—typed JSON output for reliable agent workflows.

130
TypeScript
MIT
View repository →

Auto-detects test frameworks and returns typed JSON—pytest, jest, vitest, mocha.

130
TypeScript
MIT
View repository →