#4747th of 54 in Assistants

shell_gpt

Command-line tool that generates shell commands and code from LLM prompts

Stars
12.3k
License
MIT
Last commit
Jul 2026
Last release
May 2026
Language
Python
Stars, last 29 weeks+462
21 Apr 2026: 11,847 stars28 Apr 2026: 11,872 stars5 May 2026: 11,895 stars12 May 2026: 11,918 stars19 May 2026: 11,933 stars26 May 2026: 11,956 stars2 Jun 2026: 11,979 stars9 Jun 2026: 12,004 stars16 Jun 2026: 12,015 stars23 Jun 2026: 12,030 stars30 Jun 2026: 12,048 stars7 Jul 2026: 12,077 stars14 Jul 2026: 12,100 stars21 Jul 2026: 12,106 stars28 Jul 2026: 12,123 stars4 Aug 2026: 12,150 stars11 Aug 2026: 12,162 stars18 Aug 2026: 12,174 stars25 Aug 2026: 12,194 stars1 Sept 2026: 12,207 stars8 Sept 2026: 12,217 stars15 Sept 2026: 12,233 stars22 Sept 2026: 12,237 stars29 Sept 2026: 12,248 stars6 Oct 2026: 12,256 stars13 Oct 2026: 12,266 stars20 Oct 2026: 12,278 stars27 Oct 2026: 12,287 stars3 Nov 2026: 12,301 stars10 Nov 2026: 12,309 stars

11,847 on 21 Apr 202612,309 on 10 Nov 2026

From GitHub's daily star counts.

Checked on GitHub.

Overview

ShellGPT installs as `sgpt` via pip and sends prompts from arguments or stdin to an LLM, defaulting to the OpenAI API. It generates OS- and shell-aware commands with an execute/describe/abort prompt, code-only output, named chat sessions, a REPL mode, custom roles, and function calling. Local models work through backends such as Ollama, but the README says it is not optimized for them.

Who it is for: Developers who want LLM help from the terminal

Strengths

  • Reads stdin and redirects, so it composes with pipes such as `git diff | sgpt`
  • Shell mode detects OS and `$SHELL` and asks to execute, describe or abort
  • Bash and Zsh hotkey integration puts suggested commands into the input line
  • Custom roles and Python function definitions live in `~/.config/shell_gpt`

Weaknesses

  • README says it is not optimized for local models and may not work as expected
  • Defaults to the paid OpenAI API and needs an API key
  • Function calling can run destructive shell commands; README warns to use at your own risk
  • Request cache applies only to OpenAI API requests

What it needs

  • no GPU
  • Docker
  • Needs OpenAI API key or a local backend such as Ollama, Python and pip
  • Models: OpenAI GPT models, Ollama (via LiteLLM)

Quick answers

Is shell_gpt open source?

Yes. Its code is on GitHub under the MIT license.

Can you run shell_gpt with Docker?

Yes. Its repository has a Docker setup: a Dockerfile, a Compose file or a published image.

Does shell_gpt need a GPU?

No, it runs without one.

Also in Assistants

See all 54
Also in Assistants
RankProjectScore
1ai-job-searchClaude Code workflow that searches job boards and drafts tailored CVs and cover letters71.2k starsMIT96 out of 100
2career-opsJob-search agent that scores listings, tailors CVs and drafts applications81.9k stars+8,035 (+11%) in 30 daysMIT95 out of 100
3OpenClawPersonal AI assistant reachable through Discord, Slack, Telegram, WhatsApp and more394.3k stars+4,398 (+1.1%) in 30 daysMIT93 out of 100
#4openhumanRust agent harness with desktop app, built-in memory and embeddable library42.7k stars+984 (+2.4%) in 30 daysGPL-3.093 out of 100
#5airiAI VTuber companion with VRM and Live2D avatars, voice, and game-playing Live demo ↗ (opens in a new tab)51.3k stars+1,184 (+2.4%) in 30 daysMIT91 out of 100