MCP.so
登录

Make CoPilot an Interactive Programmer MCP Server

@PEZ

基本信息

分类

其他

许可证

MIT

传输方式

stdio

作者

PEZ

关于 Make CoPilot an Interactive Programmer MCP Server

VS Code AI Agent Interactive Programming. Tools for CoPIlot and other assistants. Can also be used as an MCP server.

Make CoPilot an Interactive Programmer MCP Server 配置

暂无标准配置

该服务器的 README 中没有可解析的 MCP 配置块,请前往代码仓库查看安装说明。

代码仓库

Make CoPilot an Interactive Programmer MCP Server 工具

工具列表暂未提供

当前收录暂时没有可展示的工具列表,不代表服务没有工具。请查阅项目文档了解支持的能力。

工具是从 README 中自动提取的。维护者可以在 ## Tools 标题下列出工具,即可填充这部分内容。

概览

What is Make CoPilot an Interactive Programmer?

Make CoPilot an Interactive Programmer (Backseat Driver) is a VS Code extension that gives AI agents like GitHub Copilot and Cursor direct access to Calva’s Clojure tools. It enables the AI to evaluate code in the REPL, perform structural editing, look up symbols, and load files—turning Copilot into an interactive Clojure programmer.

How to use Make CoPilot an Interactive Programmer?

Install the Prerequisites: VS Code, the Calva extension, and the Backseat Driver extension. Then connect Calva to your Clojure/ClojureScript project. With GitHub Copilot or an MCP-compliant assistant active, ask the AI to use the REPL to investigate or implement code. For MCP-based harnesses, additional per-project configuration is required (see MCP-CONFIGURATION.md).

Key features of Make CoPilot an Interactive Programmer

  • Evaluate code via the Clojure REPL through Calva
  • Create Clojure files with automatic bracket balancing
  • Append code to files with bracket balancing
  • Structural editing (insert/replace top-level forms) with formatting, balancing, and linting
  • Bracket balancer powered by Parinfer
  • Load and evaluate entire Clojure files through the REPL
  • Symbol info lookup with doc strings and argument info
  • clojuredocs.org lookup for docs, examples, and see also

Use cases of Make CoPilot an Interactive Programmer

  • AI-assisted interactive development in Clojure/ClojureScript projects
  • Parallel subagents investigating different aspects of an implementation
  • Rapid prototyping with REPL-driven feedback directly from Copilot
  • Code refactoring with structural editing tools invoked by the AI

常见问题

1

Which AI harnesses are supported?

Copilot (zero config) and Cursor (zero config) work out of the box. Other AI harnesses can connect via MCP with some configuration per project.

2

What are the prerequisites to use the extension?

You need VS Code (or a fork), the Calva extension, the Backseat Driver extension, and either GitHub Copilot or an MCP-compliant assistant.

3

How can I limit the size of REPL evaluation results?

Set the configuration `calva-backseat-driver.evaluation.maxLength` (default 25 items) and `calva-backseat-driver.evaluation.maxDepth` (default 7 levels). Set to 0 to disable limiting.

4

Can I disable the skills that Backseat Driver provides to agents?

Yes. Use the settings `calva-backseat-driver.provideBdSkill` and `calva-backseat-driver.provideEditSkill` (both default true) to enable or disable the Backseat Driver skill and the structural editing skill.

5

Is there a UI to review AI edits?

No. Backseat Driver provides no UI for reviewing edits. Use VS Code’s source control tools and the Timeline view to inspect changes.

评论

其他 分类下的更多 MCP 服务器