@blackwell-systems/agent-lsp

中文文档
0.24.2发布于 2026/11/04MIT零依赖构建来源已验证近 30 天下载 3349

MCP server exposing 66 language server protocol tools to AI agents

原始使用文档
文档目录7 节
  1. Why
  2. Install
  3. Usage
  4. What it does
  5. Token savings
  6. Works with
  7. Links

agent-lsp

MCP server for language intelligence. Give AI agents structured access to language servers: go-to-definition, find-all-references, rename, diagnostics, completions, call hierarchy, and speculative execution.

66 tools. 24 agent workflows. 32 CI-verified languages. Single Go binary.

Why

AI coding agents navigate code with grep and file reads. This wastes 5-34x more tokens than necessary and produces 92-99% false positives on symbol lookups. Language servers already have the answers; agent-lsp makes them accessible via MCP.

Install

npm install -g @blackwell-systems/agent-lsp

The package installs the prebuilt binary for your platform. Other channels:

brew install blackwell-systems/tap/agent-lsp    # macOS/Linux
pip install agent-lsp                           # PyPI
curl -fsSL https://raw.githubusercontent.com/blackwell-systems/agent-lsp/main/install.sh | sh

Usage

# Start with auto-detection
agent-lsp

# Explicit language server
agent-lsp go:gopls typescript:typescript-language-server,--stdio

# HTTP mode (requires a bearer token)
AGENT_LSP_TOKEN=your-secret-token agent-lsp --http --port 8080 go:gopls

Then configure your AI tool's MCP settings to point at agent-lsp.

What it does

  • 66 tools covering navigation, analysis, refactoring, diagnostics, formatting, speculative execution, composite exploration, safe editing, build, test, and more
  • 24 agent workflows (skills) that encode correct multi-step operations: rename safely, analyze blast radius, simulate edits before applying
  • 32 CI-verified languages: Go, Python, TypeScript, Rust, Java, C, C++, C#, Ruby, PHP, Kotlin, Swift, Scala, Zig, Lua, Elixir, Gleam, and more
  • Speculative execution: preview edits in memory, see what breaks before touching disk
  • Phase enforcement: blocks out-of-order operations so agents follow correct workflows

Token savings

Measured across 5 codebases (Go, TypeScript, Python):

CodebaseLinesSavings
agent-lsp15K5x
Hono (TypeScript)24K13x
FastAPI (Python)33K2x
Next.js (TypeScript)196K5x
HashiCorp Consul (Go)319K34x

Full experiment: agent-lsp.com/guide/token-savings

Works with

Claude Code, Cursor, Windsurf, GitHub Copilot, and any MCP-compatible client.

版本与发布记录

@blackwell-systems/agent-lsp 共发布 47 个版本,首次发布于 2026/04/11;过去 12 个月发布了 47 个版本。最新版本附带 npm provenance 构建来源证明,可追溯到公开的 CI 构建记录。

版本发布日期距今
0.24.2latest2026/11/044 天前
0.24.12026/11/044 天前
0.24.02026/11/044 天前
0.23.62026/10/318 天前
0.23.52026/10/2712 天前
0.23.42026/10/2712 天前

关于 @blackwell-systems/agent-lsp 的常见问题

@blackwell-systems/agent-lsp 是什么?

MCP server exposing 66 language server protocol tools to AI agents

如何安装 @blackwell-systems/agent-lsp?

在项目目录运行 npm install @blackwell-systems/agent-lsp,或使用 pnpm add / yarn add / bun add。它还提供命令行工具 agent-lsp,可以用 npx agent-lsp 直接运行。

@blackwell-systems/agent-lsp 支持 TypeScript 吗?

该包的 package.json 没有声明类型入口。需要类型提示时,可以在 npm 上查找社区维护的 @types/blackwell-systems__agent-lsp 是否存在。

@blackwell-systems/agent-lsp 可以商用吗?

@blackwell-systems/agent-lsp 使用 MIT 许可。宽松许可:允许商业使用、修改和再分发,通常只需在分发时保留原版权与许可声明。以上为概括,不构成法律意见,请以许可原文为准。

@blackwell-systems/agent-lsp 的下载量有多少?

根据 npm 官方下载统计,@blackwell-systems/agent-lsp 在 2026/10/09 – 2026/11/07 这 30 天内被下载 3,349 次,日均约 112 次。下载量会把 CI 构建、镜像同步和作为其他包依赖的安装一并计入,适合横向比较热度,不能直接等同于使用人数。

@blackwell-systems/agent-lsp 还在维护吗?

最新版本 0.24.2 发布于 2026/11/04(4 天前);至今共发布 47 个版本,其中过去 12 个月发布了 47 个;首次发布于 2026/04/11。发布频率只是参考信号:功能稳定的工具可能很少需要更新,建议结合源码仓库的 issue 和提交记录判断。