# CodeSavant Coding Assistant

> Use this tool when you need to manage and execute code efficiently, solving problems in collaborative coding, code analysis, and continuous integration workflows. The CodeSavant Coding Assistant provides a robust interface for reading, writing, and versioning code files, executing shell commands, and running language-specific code. It supports multiple programming languages and offers features like diff generation, history tracking, and timeout handling, making it ideal for AI-assisted software development, code review, and automated testing scenarios.

Canonical page: https://skillsregistry.net/skills/twolven-codesavant  
JSON: https://api.skillsregistry.net/v1/skills/twolven-codesavant

## Description

This CodeSavant MCP server, developed by Todd Wolven, provides AI assistants with a robust interface for code management and execution. Built with Python and leveraging asyncio for concurrent operations, it offers capabilities for reading, writing, and versioning code files across multiple projects. The server implements a CodeFileManager for file operations, a CodeCommandExecutor for shell command execution, and a CodeRunner for language-specific code execution. It supports multiple programming languages and includes features like diff generation, history tracking, and timeout handling. This implementation is particularly valuable for AI-assisted software development, code review, and automated testing scenarios, facilitating use cases such as collaborative coding, code analysis, and continuous integration workflows.

## Trust

- **Trust score (0–1):** 0.86
- **Verification tier:** verified
- **Last scanned:** 2026-09-28

## Facts

- **Version:** 1.0.0
- **Skill type:** atomic
- **Execution layer:** mcp-remote
- **Runtime environment:** api
- **Category:** file-system
- **Updated:** 2026-09-28

## Source

- **Source listing:** [PulseMCP](https://www.pulsemcp.com/servers/twolven-codesavant)
- **Repository:** <https://github.com/twolven/mcp-codesavant>

## Use it

Resolve this record through the SkillsRegistry MCP server (no auth, read-only):

```
claude mcp add --transport http --scope user skillsregistry https://api.skillsregistry.net/mcp
```

```json
{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "get_skill",
    "arguments": {
      "slug": "twolven-codesavant"
    }
  }
}
```

REST: `GET https://api.skillsregistry.net/v1/skills/twolven-codesavant` · pull for local use: `GET https://api.skillsregistry.net/v1/skills/twolven-codesavant/pull`

---
SkillsRegistry indexes agent skills from public registries and GitHub. Skills we have analysed are scanned with Circle-IR and scored on six dimensions; each listing states its scan coverage. More: https://skillsregistry.net/llms.txt