
semantic-edit-mcp
semantic-edit-mcp
README
Semantic Edit MCP
[!CAUTION] This repository is written by AI as an experiment. Although I (jbr) review every line, the code quality is not necessarily identical to what I would have written. Caveat usor.
A Model Context Protocol (MCP) server for semantic code editing using tree-sitter. This server provides safe, AST-aware code editing operations that preserve syntax structure and formatting.
Features
- Semantic node targeting: Find nodes by name, type, tree-sitter query, or position
- Safe structural editing: Replace, insert, wrap, or delete AST nodes while maintaining syntax
- Syntax validation: Validate code before and after edits to prevent breaking changes
- Multiple languages: Currently supports Rust with extensible architecture for more languages
- Transaction safety: All edits are validated before being applied to files
Installation
cargo install semantic-edit-mcp
Usage
As an MCP Server
Start the server:
semantic-edit-mcp serve
The server communicates via JSON-RPC over stdin/stdout and provides the following tools:
Available Tools
replace_node
Replace an entire AST node with new content.
{
"file_path": "src/main.rs",
"selector": {
"type": "function_item",
"name": "main"
},
"new_content": "fn main() {\n println!(\"Hello, semantic editing!\");\n}"
}
insert_before_node
/ insert_after_node
Insert content before or after a specified node.
{
"file_path": "src/lib.rs",
"selector": {
"type": "function_item",
"name": "existing_function"
},
"content": "/// New documentation\n#[derive(Debug)]"
}
wrap_node
Wrap an existing node with new syntax.
{
"file_path": "src/lib.rs",
"selector": {
"line": 42,
"column": 10
},
"wrapper_template": "if some_condition {\n {{content}}\n}"
}
validate_syntax
Validate code syntax.
{
"file_path": "src/main.rs"
}
Or validate content directly:
{
"content": "fn test() { println!(\"test\"); }",
"language": "rust"
}
get_node_info
Get information about a node at a specific location.
{
"file_path": "src/main.rs",
"selector": {
"line": 10,
"column": 5
}
}
Node Selectors
Node selectors allow you to target specific AST nodes using different strategies:
By Position
{
"line": 42,
"column": 10
}
By Name and Type
{
"type": "function_item",
"name": "my_function"
}
By Type Only
{
"type": "struct_item"
}
By Tree-sitter Query
{
"query": "(function_item name: (identifier) @name (#eq? @name \"main\")) @function"
}
Architecture
The project is organized into several modules:
parsers/
: Tree-sitter integration and language-specific parsingeditors/
: Language-specific editing logic (currently Rust)operations/
: Core edit operations and node selectionvalidation/
: Syntax validation and error reporting
Safety Features
- Syntax Validation: All edits are validated before being applied
- AST-Aware Positioning: Edits respect semantic boundaries
- Atomic Operations: File changes are applied atomically
- Format Preservation: Maintains indentation and structure context
Extending to New Languages
To add support for a new language:
- Add the tree-sitter grammar dependency to
Cargo.toml
- Create a new parser module in
src/parsers/
- Create a new editor module in
src/editors/
- Update the language detection and dispatch logic
Examples
Replace a function with error handling
{
"name": "replace_node",
"arguments": {
"file_path": "src/main.rs",
"selector": {
"type": "function_item",
"name": "risky_operation"
},
"new_content": "fn risky_operation() -> Result<(), Box<dyn Error>> {\n // Safe implementation\n Ok(())\n}"
}
}
Add documentation to a struct
{
"name": "insert_before_node",
"arguments": {
"file_path": "src/lib.rs",
"selector": {
"type": "struct_item",
"name": "MyStruct"
},
"content": "/// A well-documented struct\n/// \n/// This struct represents..."
}
}
Wrap code in a conditional
{
"name": "wrap_node",
"arguments": {
"file_path": "src/main.rs",
"selector": {
"line": 25,
"column": 4
},
"wrapper_template": "#[cfg(feature = \"advanced\")]\n{{content}}"
}
}
License
MIT OR Apache-2.0
推荐服务器

Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。

VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。

e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。