omp — AI coding agent viết bằng Rust (27K lines), hash-anchored edits (-61% output tokens), LSP+debugger tích hợp, 32 tools, 40+ LLM providers. Fork của Pi platform.
Scanned 9/9/2026
Install to Claude Code
npx -y skills add yanacuti1121/Yana-AI --skill can1357--oh-my-pi --agent claude-codeInstalls into .claude/skills of the current project.
Are you the author of Can1357 Oh My Pi?
Add the live security badge to your README — it updates automatically with every re-scan.
[](https://www.skillsdirectory.com/skills/yanacuti1121-can1357-oh-my-pi)More formats (shields.io, HTML) on the badges page.
---
name: can1357--oh-my-pi
description: "omp — AI coding agent viết bằng Rust (27K lines), hash-anchored edits (-61% output tokens), LSP+debugger tích hợp, 32 tools, 40+ LLM providers. Fork của Pi platform."
allowed-tools: Bash, Read, Write
user-invocable: true
---
omp (oh-my-pi): coding agent terminal-first với LSP và debugger wired in — không phải wrapper của IDE, mà là agent hiểu code như IDE hiểu.
## Install
```bash
# macOS/Linux
curl -fsSL https://omp.sh/install -o /tmp/omp-install.sh
# Inspect first: head -40 /tmp/omp-install.sh — then run if safe:
sh /tmp/omp-install.sh
# Bun
bun install -g @oh-my-pi/pi-coding-agent
# Windows
irm https://omp.sh/install.ps1 | iex
```
## Core Innovation: Hash-Anchored Edits
```
Thay vì: model rewrite dòng code (nhiều tokens, dễ lỗi whitespace)
omp dùng: content-hash anchors → model CHỈ point tới anchor + thay đổi cần làm
Kết quả: -61% output tokens trên same task
```
## LSP Integration
```
Rename symbol → omp route qua workspace/willRenameFiles
→ tự động update re-exports, barrel files, aliased imports
→ agent biết mọi thứ IDE biết: go-to-definition, references, diagnostics
```
## Debugger Support
```
Supports: lldb (C/C++/Rust), dlv (Go), debugpy (Python)
- Attach to segfaulting binaries
- Step through code, inspect frames
- Analyze goroutines in hanging services
```
## 32 Tools (8 categories)
```
File: read, write, edit, ast_edit, search
Runtime: bash, eval, ssh
Code: lsp, debug
Coord: task, irc, todo
External: browser, web_search, github
Memory: retain, recall, reflect
```
## Multi-Provider (40+)
```
Anthropic, OpenAI, Gemini, xAI Grok, Mistral, Ollama, llama.cpp
Role-based routing:
default → standard turns
smol → cost-effective subagents
slow → deep reasoning
plan → strategic planning
```
## Subagents trong isolated worktrees
```typescript
// spawn subagent, results là schema-validated objects
const result = await agent.task({
prompt: "analyze this module",
worktree: true // isolated, no merge conflicts
})
```
## Import config tự động
Tự nhận: Cursor MDC, Cline rules, Codex AGENTS.md, Copilot settings — không cần migrate.
## Source
https://github.com/can1357/oh-my-pi · MIT · +2318⭐/week
Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.
No comments yet. Be the first to comment!