惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

B
Blog RSS Feed
博客园 - 叶小钗
F
Fortinet All Blogs
GbyAI
GbyAI
Martin Fowler
Martin Fowler
博客园 - 聂微东
I
InfoQ
B
Blog
IT之家
IT之家
美团技术团队
L
LangChain Blog
小众软件
小众软件
C
Check Point Blog
MongoDB | Blog
MongoDB | Blog
Cyber Security Advisories - MS-ISAC
Cyber Security Advisories - MS-ISAC
V
V2EX
Last Week in AI
Last Week in AI
A
About on SuperTechFans
博客园 - Franky
P
Proofpoint News Feed
罗磊的独立博客
月光博客
月光博客
V
Visual Studio Blog
MyScale Blog
MyScale Blog

Show HN

Show HN: AI agents for UK GDAD PCF roles and their skills The Two Pillars: Mixer Mode and Meta-Software in the Reorganization of Software Work After AI GitHub - JaiCode08/teleport-env What 1,000+ Harness Experiments Taught Me About Self-Improving Agents Show HN: Liiists, a Markdown-first, iOS and CLI list app SwiperTab – Get this Extension for 🦊 Firefox (en-US) GitHub - kouhxp/fftext: Summarize, explain, fact-check, or translate any text, URL, or file. No GPU. No cloud. One command GitHub - sweetpad-dev/sweetpad: Develop Swift/iOS projects using VSCode GitHub - dogmaticdev/IRON: IRON a.k.a. Intermediate Representation Object Notation is a Interpreter/Database that is used to create Programming Languages. GitHub - sjhalani7/vaen: Package your AI coding harness into a portable .agent file, and share it across repos, teams, & the community without ever having to copy-paste instructions, skills, MCP config, or secrets. Show HN: Gandalf the Grader Show HN: Citadeld – replay any CI failure locally from a single file GitHub - tdortman/cuSBF: High-Performance GPU Super Bloom Filter coral-ai/claude-code-token-xray at main · Coral-Bricks-AI/coral-ai GitHub - ulyssestenn/funes: Funes is a Git-based framework for LLM-managed knowledge work: an AI Librarian ingests raw sources, builds an interlinked Markdown knowledge base, and uses it to produce cited reports, analyses, and other outputs. GitHub - ThatXliner/gah: Git Add Hunk, built for agents to use GitHub - harmont-dev/harmont-cli: Command-line client for the Harmont CI platform GitHub - brooksmcmillin/mcp-authflow: OAuth 2.0 Authorization Server framework for MCP servers GitHub - javaid-codes/audit-supply-chain-agents GitHub - amorey/gochan: A small library of common channel architectures for Go, inspired by Rust GitHub - arifozgun/OpenGem: Free, Open-Source AI API Gateway with Gemini, OpenAI & Anthropic Compatibility in 1 file GitHub - Pranesh950/BioPetals: 🌸 Run BIOxAI models at home, BitTorrent-style. Fine-tuning and inference up to 10x faster than offloading GitHub - cnguyen14/bounty-doctor: Diagnose a GitHub bounty issue before you waste hours: detects honeypot scam repos, AI-bot attempt swarms, and stale contests. Show HN: CoreMCP – MCP Server for On-Prem DBs Show HN: KittyHTML – Render HTML/CSS as an inline image in your terminal GitHub - bingud/filemat: Web-based file manager Show HN: TruthLens – Free multi-signal deepfake image detector GitHub - apexlocal-jz/claude-usage-tray: Windows system-tray app showing your Claude Code rate-limit usage at a glance. Zero deps, ~300 lines of PowerShell. Cross-IDE (works regardless of VS Code, Cursor, plain terminal). Release v0.1.2.1 · kouhxp/yapsnap GitHub - noopolis/moltnet: Self-hostable chat network for AI agents. Pre-built bridges for Claude Code, Codex, and the Claws. Rooms, DMs, history. No Slack bots, no Matrix, no glue code.
GitHub - longyijdos/hi-shell: AI-powered zsh command gene...
longyijdos · 2026-06-20 · via Show HN

hi-shell logo

AI-powered zsh command generation with reviewable, risk-scored shell suggestions.

hi-shell demo

hi-shell turns natural language into shell commands and shows the result as ghost text in your current zsh line. Press Tab to accept the suggestion, edit it if needed, then press Enter yourself.

It is not an autonomous shell agent, terminal emulator, or chat UI. It keeps command generation inside your normal terminal workflow while leaving execution under your control.

✨ Why hi-shell

Shell commands are often easy to describe and annoying to spell out. hi-shell handles the translation without taking over the terminal.

  • Stay in zsh instead of switching to a chat window.
  • Review every generated command before it can run.
  • Accept suggestions into the real command line, not a separate UI.
  • Revise a suggestion or ask what it does before running it.
  • Block clearly catastrophic commands by default.

⚡ Features

  • Natural-language command generation from your shell.
  • Ghost-text suggestions powered by zsh ZLE widgets.
  • Tab-to-accept flow; Enter only runs after you accept.
  • safe, warn, and blocked risk scoring.
  • Built-in support for OpenAI and DeepSeek APIs.
  • OpenAI-compatible provider configuration.
  • Optional shell context and filtered history.
  • Clean install and uninstall with one managed .zshrc block.

🚀 Install

curl -fsSL https://raw.githubusercontent.com/longyijdos/hi-shell/main/scripts/install.sh | sh
exec zsh

Make sure ~/.local/bin is in your PATH.

From source:

git clone https://github.com/longyijdos/hi-shell.git
cd hi-shell
./scripts/install.sh
exec zsh

Prerequisites for source installs: Go 1.22+, git, and zsh.

⚙️ Configure

OpenAI is the default provider:

export OPENAI_API_KEY="sk-..."

hi-shell config set provider openai
hi-shell config set openai.api_key_env OPENAI_API_KEY
hi-shell config set openai.model gpt-4.1-mini

DeepSeek is also supported:

export DEEPSEEK_API_KEY="sk-..."

hi-shell config set provider deepseek
hi-shell config set deepseek.api_key_env DEEPSEEK_API_KEY
hi-shell config set deepseek.model deepseek-v4-flash

Secrets stay in environment variables. The config file stores environment variable names, not API keys.

See Configuration for the full config reference.

🧭 Use

Type a natural-language request with the hi prefix:

Press Enter to generate a suggestion. If the command looks right, press Tab to accept it into your shell input line. Press Enter again to run it.

Use the prefix key, Ctrl-] by default, to revise, ask about, edit, or dismiss the current suggestion. See Usage and Keybindings for details.

🛡️ Safety

hi-shell never runs generated commands automatically.

The shell plugin only inserts a suggestion after you accept it. You still review the final command and press Enter yourself. Local risk scoring classifies suggestions as safe, warn, or blocked; clearly catastrophic commands such as rm -rf / are blocked by default.

See Risk Scoring for the detailed model.

📚 Documentation

🧹 Uninstall

hi-shell uninstall
exec zsh

To remove all hi-shell files, including config:

hi-shell uninstall --purge
exec zsh

📄 License

MIT