deepy-cli 0.1.8

Deepy - Vibe coding for DeepSeek models in your terminal

Deepy

A cute terminal coding agent for DeepSeek models.
Read, edit, run tools, search the web, and keep project context in one terminal session.

Website · 中文文档 · Quick Start

What Is Deepy?

Deepy is a Python terminal coding agent built for DeepSeek's OpenAI-compatible models. It keeps the workflow inside your terminal: ask questions, inspect a project, edit files, run commands, search or fetch web content, and resume the same project session later.

Deepy is designed around DeepSeek V4 thinking mode, long context, cache-friendly prompting, and a Rich terminal interface that makes tool calls, diffs, usage, and context state visible while the agent works.

Highlights

• DeepSeek-first model setup with deepseek-v4-pro, thinking enabled, and reasoning_effort=max by default; /model can switch between V4 Pro and V4 Flash with none, high, or max thinking strength.
• OpenAI Agents SDK integration through OpenAIChatCompletionsModel.
• Project-aware coding tools for reading files, modifying files, running shell commands, and showing readable diffs.
• Web research tools for search and direct URL fetching when a task needs fresh information.
• Session history, /resume, /new, automatic context tracking, and compacting for long project work.
• TOML-only private configuration at ~/.deepy/config.toml.
• Theme-aware terminal UI with Markdown rendering, DeepSeek thinking display, per-turn usage, context window status, and version update checks.

See It Work

Start In A Project

Deepy shows the current model, thinking settings, working directory, and the core commands directly on startup.

Build And Verify Code

Ask Deepy to implement a change, write tests, run the project test command, and summarize the result.

Deepy can also turn command output into a readable project summary, including files created, code snippets, and test coverage.

Research With Sources

Deepy includes WebSearch and WebFetch tools, so a terminal session can gather current information and fetch exact pages when a URL is provided.

Quick Start

Install from PyPI after the first release:

uv tool install deepy-cli

The installed command is still deepy.

Install the latest code from GitHub:

uv tool install git+https://github.com/kirineko/deepy.git

Configure your DeepSeek API key:

deepy config setup

Start Deepy in a project:

cd your-project
deepy

Configuration

Deepy only uses TOML configuration. JSON config files are intentionally rejected.

# ~/.deepy/config.toml
[model]
api_key = "sk-..."
name = "deepseek-v4-pro"
base_url = "https://api.deepseek.com"
thinking = true
reasoning_effort = "max" # high or max when thinking is enabled

[context]
window_tokens = 1048576
compact_trigger_ratio = 0.8
reserved_context_tokens = 50000
compact_preserve_recent_messages = 2

[ui]
theme = "auto" # auto, dark, or light

Supported interactive model choices are deepseek-v4-pro and deepseek-v4-flash. In /model, thinking strength none saves thinking = false; high and max save thinking = true with the matching reasoning_effort.

WebSearch uses Deepy's hosted SearXNG endpoint by default. You can override it with your own SearXNG instance:

[tools.web_search]
searxng_url = "https://your-searxng.example/"

You can also initialize config non-interactively:

deepy config init --api-key sk-... --model deepseek-v4-pro

If your terminal uses a light background and parts of the UI look low contrast, set the UI theme explicitly:

deepy config theme light

Common Commands

deepy --version
deepy config setup
deepy config reset
deepy config theme
deepy config theme light
deepy doctor
deepy doctor --live --json
deepy status
deepy skills list
deepy sessions list
deepy sessions show <session-id>
deepy run "summarize this project"

Inside the interactive terminal:

/skills   List available skills
/model    Select model and thinking strength
/new      Start a fresh conversation
/resume   Pick a previous session
/compact  Compact the active session context
/theme    Show or change UI theme
/reset    Delete config and run setup again
/         Open the command menu
Esc       Interrupt the current model turn
Ctrl+D    Press twice to quit

Project Rules And Skills

Deepy automatically loads project instructions from:

• AGENTS.md
• .deepy/skills/*/SKILL.md

This lets each repository define local conventions, commands, review rules, and domain-specific skills without changing global config.

Development

uv sync --group dev
uv run pytest
uv run ruff check
uv run pyright
uv build

The Python package is built from src/deepy. GitHub Pages files and screenshot assets live outside the package directory and are not included in the wheel.

Release Status

Deepy 0.1.8 is released through GitHub and PyPI. Standalone binaries and npm wrappers can be added later, but the primary distribution is the Python CLI.