
Aicoreutils
- 1 repo stars
- Updated June 22, 2026
- caseSHY/AI-CLI
AI Coreutils is a MCP server that exposes 114 JSON-first coreutils commands with sandbox and dry-run for AI coding agents.
About
AI Coreutils (aicoreutils) is a PyPI-distributed CLI and MCP server that reimagines classic Unix coreutils for AI agents: commands emit JSON, support sandboxing and dry-run, and cover a broad catalog of 114 operations described in the registry manifest. developers who rely on Claude Code or Codex to refactor repositories, normalize files, or orchestrate local automation get a safer alternative to unconstrained shell access—the agent can preview effects, operate inside a sandbox, and parse structured results. Skillselion indexes it under Build → Agent-tooling because it augments the agent’s hands during implementation and ops scripts, not as a replacement for formal test suites or production monitoring stacks. It pairs well with git-centric workflows and custom skills that need deterministic file operations; treat sandbox boundaries and host permissions as your responsibility when moving from dry-run to live execution.
- 114 JSON-first Coreutils-style commands exposed for programmatic agent use
- Sandbox and dry-run modes to reduce destructive mistakes during agent-driven edits
- Dual surface: CLI plus MCP (PyPI package aicoreutils, manifest version 0.3.6)
- stdio transport via PyPI install—repository caseSHY/AI-CLI on GitHub
- Suited to repo hygiene, batch file transforms, and scripted agent workflows on a solo dev machine
Aicoreutils by the numbers
- Data as of Jul 7, 2026 (Skillselion catalog sync)
claude mcp add aicoreutils -- uvx aicoreutilsAdd your badge
Show developers this MCP server is listed on Skillselion. Paste this into your README.
| repo stars | ★ 1 |
|---|---|
| Package | aicoreutils |
| Transport | STDIO |
| Auth | None |
| Last updated | June 22, 2026 |
| Repository | caseSHY/AI-CLI ↗ |
What it does
Let agents run sandboxed, JSON-first Unix-style file and shell operations through 114 coreutils commands with dry-run guardrails.
Who is it for?
Power users automating repo maintenance and local DevOps with agents who want guardrailed coreutils instead of ad-hoc bash.
Skip if: Beginners who only need IDE formatting, or teams seeking hosted cloud infra MCP rather than on-machine CLI tools.
What you get
After installing aicoreutils and registering the stdio MCP server, agents execute cataloged commands with JSON output, sandbox, and dry-run controls.
- MCP-accessible catalog of 114 JSON-first utility commands
- Sandboxed and dry-run pathways for agent-driven filesystem work
- Repeatable local automation without hand-written shell wrappers
By the numbers
- 114 commands documented in registry description
- Manifest version 0.3.6; PyPI package identifier aicoreutils
- Transport: stdio
README.md
AICoreUtils
Glama 92% | TDQS A 级 (均值 4.6) | 114 工具全部 A 级 | CI 全平台通过 | Production/Stable
🤖 MCP 目录已收录:Glama · ModelScope · awesome-mcp-servers
中文说明
AICoreUtils 是一个面向 LLM Agent 的 JSON 优先命令行工具包原型。它参考 GNU Coreutils 的常用命令,但不是完整的 GNU 兼容替代品。
项目目标是给机器调用方提供确定、低噪音、易解析的 CLI 接口:
- 默认输出 JSON
- 错误以 JSON 写入 stderr
- 退出码语义稳定
- 修改文件的命令支持
--dry-run - 需要管道组合时显式使用
--raw
快速开始
pip install aicoreutils
aicoreutils schema --pretty
aicoreutils ls . --limit 20
aicoreutils rm build --recursive --dry-run
🤖 Claude Desktop / MCP 集成
一行配置,让 Claude 直接操作你的文件系统:
编辑 Claude Desktop 配置文件(详细说明 →):
| 系统 | 配置文件 |
|---|---|
| macOS | ~/Library/Application Support/Claude/claude_desktop_config.json |
| Windows | %APPDATA%\Claude\claude_desktop_config.json |
| Linux | ~/.config/Claude/claude_desktop_config.json |
{
"mcpServers": {
"aicoreutils": {
"command": "python",
"args": ["-m", "aicoreutils.mcp_server"]
}
}
}
重启 Claude Desktop,然后对它说:
"列出项目里所有 Python 文件,统计代码行数"
Claude 自动调用 aicoreutils ls + aicoreutils wc,全程 JSON 交互。
更多集成方式:aicoreutils tool-list --format openai 输出 OpenAI Function Calling 格式,可直接用于任意 Agent 框架。
如需给调度器或审计系统保留风险标签,可追加 --include-risk。
⚠️ 安全提示:生产环境建议以最低权限运行。
aicoreutils-mcp --profile readonly # 推荐:只读工具 aicoreutils-mcp --profile workspace-write # 仅允许低风险 cwd 内写入详见 生产安全部署指南 →
🤖 AI IDE 集成
在 Cursor / Windsurf / Continue.dev 中直接使用 aicoreutils:AI IDE 集成指南 →
// ~/.cursor/mcp.json
{ "mcpServers": { "aicoreutils": { "command": "python", "args": ["-m", "aicoreutils.mcp_server"] } } }
🔗 更多:Claude Desktop 集成 | AI IDE 集成 | Agent 任务示例 | LangChain 包装器
运行测试
# 推荐主入口(pytest,含 Hypothesis property-based 测试和 GNU 对照测试)
uv run pytest tests/ -v --tb=short
# Legacy 入口(unittest,部分运行器)
uv run python -m unittest discover -s tests -v
项目结构
.
|-- src/aicoreutils/ # Python 包源码
|-- docs/ # 文档入口
| |-- reference/ # 协议、命令面和安全生产契约
| |-- guides/ # 使用指南
| |-- architecture/ # 架构决策记录 (ADR) 和 AI 代理规则
| |-- development/ # 测试和开发说明
| |-- status/ # 当前项目状态(唯一权威来源)
| |-- audits/ # 兼容性和质量审计
| |-- analysis/ # 项目分析日志(历史归档)
| `-- reports/ # 测试报告等生成/归档文档
|-- tests/ # 测试套件
|-- examples/ # 示例
|-- scripts/ # CI/审计/发布脚本
|-- .github/ # CI workflows 和 issue 模板
`-- vendor/ # 本地上游源码缓存
文档
发布状态
当前实现:aicoreutils schema 中登记 114 个 CLI 命令(含 tool-list 等 Agent 元命令)。
重要限制:本项目是受 GNU Coreutils 启发的 Agent 友好子集,不是完整的 GNU Coreutils 克隆。
English
AICoreUtils is a JSON-first command-line toolkit prototype for LLM agents. It is inspired by common GNU Coreutils commands, but it is not a complete GNU-compatible replacement.
The goal is a deterministic, low-noise interface for machine callers:
- JSON output by default
- JSON errors on stderr
- Stable semantic exit codes
--dry-runfor mutation commands- Explicit
--rawoutput for pipeline composition
Quick Start
pip install aicoreutils
aicoreutils schema --pretty
aicoreutils ls . --limit 20
aicoreutils rm build --recursive --dry-run
🤖 Claude Desktop / MCP Integration
One config line to let Claude operate your filesystem:
Edit Claude Desktop config (full guide →):
| OS | Config File |
|---|---|
| macOS | ~/Library/Application Support/Claude/claude_desktop_config.json |
| Windows | %APPDATA%\Claude\claude_desktop_config.json |
| Linux | ~/.config/Claude/claude_desktop_config.json |
{
"mcpServers": {
"aicoreutils": {
"command": "python",
"args": ["-m", "aicoreutils.mcp_server"]
}
}
}
Restart Claude Desktop, then ask:
"List all Python files in the project and count lines of code"
Claude calls aicoreutils ls + aicoreutils wc automatically.
For other frameworks: aicoreutils tool-list --format openai outputs OpenAI Function Calling format directly.
Add --include-risk when an orchestrator or audit system needs machine-readable risk metadata.
⚠️ Security: Run with least privilege in production.
aicoreutils-mcp --profile readonly # Recommended: read-only tools aicoreutils-mcp --profile workspace-write # Low-risk cwd-local writes only
Run tests
# Recommended primary entry (pytest, includes Hypothesis property-based and GNU differential tests)
uv run pytest tests/ -v --tb=short
# Legacy entry (unittest, partial runner)
uv run python -m unittest discover -s tests -v
Project Layout
.
|-- src/aicoreutils/ # Python package
|-- docs/ # documentation index
| |-- reference/ # protocol, command-surface and security contracts
| |-- guides/ # usage guides
| |-- architecture/ # ADRs and AI agent governance rules
| |-- development/ # testing and development notes
| |-- status/ # current project status (single authoritative source)
| |-- audits/ # compatibility and quality audits
| |-- analysis/ # project analysis logs (historical archive)
| `-- reports/ # test reports and archived generated docs
|-- tests/ # test suite
|-- examples/ # examples
|-- scripts/ # CI/audit/release scripts
|-- .github/ # CI workflows and issue templates
`-- vendor/ # local upstream source cache
Documentation
- Documentation index
- Current project status ← authoritative status source
- Agent protocol and examples
- Security model
- Chinese/English user guide
- GNU Coreutils compatibility audit
- Testing guide
- WSL local CI
- Documentation governance rules
- Fact propagation matrix
Release Status
Current implementation: 114 CLI commands in aicoreutils schema (including agent-native meta-commands like tool-list).
Important limitation: this project is an agent-friendly subset inspired by GNU Coreutils, not a full GNU Coreutils clone.
稳定性和 SemVer
aicoreutils 从 v1.0.0 起采用语义化版本控制,承诺如下:
- Patch (1.0.x):修复 bug、改进错误消息、补充文档。JSON 输出结构不变。
- Minor (1.x.0):新增命令、新增参数。已有命令的 JSON 输出结构保持向后兼容。
- Major (x.0.0):破坏性变更 — JSON schema 变化、命令重命名、MCP tool schema 变化。
⚠️ Stability note: JSON envelope (ok, result, error, command, version), MCP tool schema, and semantic exit codes are stable. Production use: pin the version (pip install aicoreutils==1.2.3). v1.2.3 LTS — critical bug and security fixes backported for at least 12 months. CLI internal argument parsing may evolve across minor versions. See Stability & SemVer.
Recommended MCP Servers
How it compares
Sandboxed agent-oriented coreutils MCP, not a cloud deploy server or a web-scraping browser automation skill.
FAQ
Who is AI Coreutils MCP for?
Developers and agent-heavy developers who want JSON-structured file and utility commands with sandbox and dry-run when Claude Code, Cursor, or Codex touches the filesystem.
When should I use AI Coreutils MCP?
Use it during Build when scripting agent-driven refactors, batch file tasks, or local automation—especially when you need dry-run before destructive changes.
How do I add AI Coreutils MCP to my agent?
Install the PyPI package aicoreutils (version aligned with the MCP manifest), configure the stdio MCP entry in your client, and ensure Python runtime requirements from the caseSHY/AI-CLI repository are met.