@duheso/zerocli

Zero CLI

Zero CLI is an open-source coding-agent CLI for cloud and local model providers.

Use OpenAI-compatible APIs, Gemini, GitHub Models, Codex OAuth, Codex, Ollama, Atomic Chat, and other supported backends while keeping one terminal-first workflow: prompts, tools, agents, MCP, slash commands, and streaming output.

Quick Start | Setup Guides | Tribunal do Conselho | Providers | Source Build | VS Code Extension | Community

Why Zero CLI

• Use one CLI across cloud APIs and local model backends
• Save provider profiles inside the app with /provider
• Run with OpenAI-compatible services, Gemini, GitHub Models, Codex OAuth, Codex, Ollama, Atomic Chat, and other supported providers
• Keep coding-agent workflows in one place: bash, file tools, grep, glob, agents, tasks, MCP, and web tools
• Invoke the Tribunal do Conselho to have 5 AI advisors debate any problem from radically different perspectives
• Use the bundled VS Code extension for launch integration and theme support

Quick Start

Prerequisites

Node.js 20 or later is required. Verify with node --version.

If your distro ships an older Node.js (common on AlmaLinux, CentOS, Fedora, RHEL), upgrade first:

curl -o- https://raw.githubusercontent.com/nvm-sh/nvm/v0.40.3/install.sh | bash
source ~/.bashrc          # or ~/.zshrc / ~/.profile
nvm install 22
nvm use 22
node --version            # should print v22.x.x

# Fedora / RHEL 9+ / AlmaLinux 9+
curl -fsSL https://rpm.nodesource.com/setup_22.x | sudo bash -
sudo dnf install -y nodejs

Install

npm install -g @duheso/zerocli

If the install later reports ripgrep not found, install ripgrep system-wide and confirm rg --version works in the same terminal before starting Zero CLI.

Start

zero

Inside Zero CLI:

• run /provider for guided provider setup and saved profiles
• run /onboard-github for GitHub Models onboarding

Fastest OpenAI setup

macOS / Linux:

export CLAUDE_CODE_USE_OPENAI=1
export OPENAI_API_KEY=sk-your-key-here
export OPENAI_MODEL=gpt-4o

zero

Windows PowerShell:

$env:CLAUDE_CODE_USE_OPENAI="1"
$env:OPENAI_API_KEY="sk-your-key-here"
$env:OPENAI_MODEL="gpt-4o"

zero

Fastest local Ollama setup

macOS / Linux:

export CLAUDE_CODE_USE_OPENAI=1
export OPENAI_BASE_URL=http://localhost:11434/v1
export OPENAI_MODEL=qwen2.5-coder:7b

zero

Windows PowerShell:

$env:CLAUDE_CODE_USE_OPENAI="1"
$env:OPENAI_BASE_URL="http://localhost:11434/v1"
$env:OPENAI_MODEL="qwen2.5-coder:7b"

zero

Using COPILOT_* env vars (remote or local server)

The COPILOT_* variables are an alternative mapping supported by the launcher. They are translated to the native OpenAI-compatible vars at startup, so you never need to set both.

Remote server — macOS / Linux:

export COPILOT_PROVIDER_BASE_URL=http://10.11.36.131:3053/v1
export COPILOT_PROVIDER_TYPE=openai
export COPILOT_PROVIDER_API_KEY=not-required
export COPILOT_MODEL=gpt-oss-120b

zero

Remote server — Windows PowerShell:

$env:COPILOT_PROVIDER_BASE_URL="http://10.11.36.131:3053/v1"
$env:COPILOT_PROVIDER_TYPE="openai"
$env:COPILOT_PROVIDER_API_KEY="not-required"
$env:COPILOT_MODEL="gpt-oss-120b"

zero

Localhost — macOS / Linux:

export COPILOT_PROVIDER_BASE_URL=http://localhost:3053/v1
export COPILOT_PROVIDER_TYPE=openai
export COPILOT_PROVIDER_API_KEY=not-required
export COPILOT_MODEL=gpt-oss-120b

zero

Localhost — Windows PowerShell:

$env:COPILOT_PROVIDER_BASE_URL="http://localhost:3053/v1"
$env:COPILOT_PROVIDER_TYPE="openai"
$env:COPILOT_PROVIDER_API_KEY="not-required"
$env:COPILOT_MODEL="gpt-oss-120b"

zero

Note: COPILOT_* vars are only applied when the matching native var (OPENAI_BASE_URL, etc.) is not already set, so the two styles can coexist without conflict.

Using Ollama's launch command

If you have Ollama installed, you can skip the env var setup entirely:

ollama launch zero --model qwen2.5-coder:7b

This automatically sets ANTHROPIC_BASE_URL, model routing, and auth so all API traffic goes through your local Ollama instance. Works with any model you have pulled — local or cloud.

Setup Guides

Beginner-friendly guides:

• Non-Technical Setup
• Windows Quick Start
• macOS / Linux Quick Start

Advanced and source-build guides:

• Advanced Setup
• Android Install

Supported Providers

What Works

• Tool-driven coding workflows: Bash, file read/write/edit, grep, glob, agents, tasks, MCP, and slash commands
• Streaming responses: Real-time token output and tool progress
• Tool calling: Multi-step tool loops with model calls, tool execution, and follow-up responses
• Images: URL and base64 image inputs for providers that support vision
• Provider profiles: Guided setup plus saved .zerocli-profile.json support
• Local and remote model backends: Cloud APIs, local servers, and Apple Silicon local inference
• Tribunal do Conselho: Multi-agent deliberation with 5 AI advisors modeled on historical figures

Tribunal do Conselho — Council Tribunal

The Tribunal do Conselho is a signature Zero CLI skill that convenes a panel of 5 AI advisors to debate any question, dilemma, or decision from radically different perspectives. Rather than getting a single AI opinion, you get a full deliberation — with agreements, disagreements, and a final synthesized verdict.

This is invaluable for complex decisions, brainstorming, code architecture review, life choices, or any situation where a single perspective is not enough.

How to invoke

/council  Should I leave my job and start a company?
/council  Review this architecture — microservices vs monolith for a 3-person team
/council  I'm getting married in 6 months. How should I prepare financially?

Aliases also work: /conselho and /tribunal.

Or just type naturally and the model will invoke it:

invoco o tribunal do conselho — quero largar meu emprego

How it works

User prompt
    │
    ▼
📨 Mensageiro (Messenger)
    Analyzes complexity → generates 3-13 clarifying questions
    │
    ▼
❓ AskUserQuestion (per question)
    Collects all context from the user
    │
    ▼
⚖️ TRIBUNAL DO CONSELHO — Round 1 / 2 / 3
    5 advisors deliberate in parallel
    React to each other, agree, disagree, debate
    │
    ▼
🔍 Convergence check after Round 3
    ├── Consensus reached → proceed to verdict
    └── Still diverging → ⚠️ 2 extra rounds invoked
                             "DIVERGÊNCIA NO CONSELHO — Rodadas extras invocadas!"
    │
    ▼
📜 Síntese — Synthesizer consolidates the verdict
    │
    ▼
✅ Final verdict delivered to the user

The Classic Council (default pack)

Five legendary figures with radically different worldviews ensure every angle of your problem is covered:

Council Packs

Switch the deliberating panel by passing pack: in your prompt or using the tool parameter:

Dynamic rounds

The debate starts with 3 rounds (minimum). After round 3, convergence is automatically detected:

• Consensus reached → debate closes, verdict synthesized immediately
• Significant divergence → a bold visual alert appears and 2 extra rounds are invoked:

⚠️  DIVERGÊNCIA NO CONSELHO — Rodadas extras invocadas!
🔥  O Tribunal detectou divergências significativas. Invocando 2 rodadas extras finais...

Dynamic clarification questions

The Messenger automatically calibrates the number of clarifying questions (3–13) based on the detail level of your prompt:

• Vague / short prompt → 8–13 questions to fill all gaps
• Moderate detail → 5–7 focused questions
• Detailed prompt → 3–5 targeted questions on remaining blind spots

Live progress streaming

You see the debate unfold in real time as each round completes — no waiting until the end to see results. Each advisor's response streams as it finishes, round by round.

Zero CLI supports multiple providers, but behavior is not identical across all of them.

• Anthropic-specific features may not exist on other providers
• Tool quality depends heavily on the selected model
• Smaller local models can struggle with long multi-step tool flows
• Some providers impose lower output caps than the CLI defaults, and Zero CLI adapts where possible

For best results, use models with strong tool/function calling support.

Agent Routing

Zero CLI can route different agents to different models through settings-based routing. This is useful for cost optimization or splitting work by model strength.

Add to ~/.zerocli/settings.json:

{
  "agentModels": {
    "deepseek-chat": {
      "base_url": "https://api.deepseek.com/v1",
      "api_key": "sk-your-key"
    },
    "gpt-4o": {
      "base_url": "https://api.openai.com/v1",
      "api_key": "sk-your-key"
    }
  },
  "agentRouting": {
    "Explore": "deepseek-chat",
    "Plan": "gpt-4o",
    "general-purpose": "gpt-4o",
    "frontend-dev": "deepseek-chat",
    "default": "gpt-4o"
  }
}

When no routing match is found, the global provider remains the fallback.

Note: api_key values in settings.json are stored in plaintext. Keep this file private and do not commit it to version control.

Web Search and Fetch

By default, WebSearch works on non-Anthropic models using DuckDuckGo. This gives GPT-4o, DeepSeek, Gemini, Ollama, and other OpenAI-compatible providers a free web search path out of the box.

Note: DuckDuckGo fallback works by scraping search results and may be rate-limited, blocked, or subject to DuckDuckGo's Terms of Service. If you want a more reliable supported option, configure Firecrawl.

For Anthropic-native backends and Codex responses, Zero CLI keeps the native provider web search behavior.

WebFetch works, but its basic HTTP plus HTML-to-markdown path can still fail on JavaScript-rendered sites or sites that block plain HTTP requests.

Set a Firecrawl API key if you want Firecrawl-powered search/fetch behavior:

export FIRECRAWL_API_KEY=your-key-here

With Firecrawl enabled:

• WebSearch can use Firecrawl's search API while DuckDuckGo remains the default free path for non-ZeroCLI models
• WebFetch uses Firecrawl's scrape endpoint instead of raw HTTP, handling JS-rendered pages correctly

Free tier at firecrawl.dev includes 500 credits. The key is optional.

Headless gRPC Server

Zero CLI can be run as a headless gRPC service, allowing you to integrate its agentic capabilities (tools, bash, file editing) into other applications, CI/CD pipelines, or custom user interfaces. The server uses bidirectional streaming to send real-time text chunks, tool calls, and request permissions for sensitive commands.

1. Start the gRPC Server

Start the core engine as a gRPC service on localhost:50051:

npm run dev:grpc

Configuration

2. Run the Test CLI Client

We provide a lightweight CLI client that communicates exclusively over gRPC. It acts just like the main interactive CLI, rendering colors, streaming tokens, and prompting you for tool permissions (y/n) via the gRPC action_required event.

In a separate terminal, run:

npm run dev:grpc:cli

Note: The gRPC definitions are located in src/proto/zero.proto. You can use this file to generate clients in Python, Go, Rust, or any other language.

Source Build And Local Development

bun install
bun run build
node dist/cli.mjs

Helpful commands:

• bun run dev
• bun test
• bun run test:coverage
• bun run security:pr-scan -- --base origin/main
• bun run smoke
• bun run doctor:runtime
• bun run verify:privacy
• focused bun test ... runs for the areas you touch

Testing And Coverage

Zero CLI uses Bun's built-in test runner for unit tests.

Run the full unit suite:

bun test

Generate unit test coverage:

bun run test:coverage

Open the visual coverage report:

open coverage/index.html

If you already have coverage/lcov.info and only want to rebuild the UI:

bun run test:coverage:ui

Use focused test runs when you only touch one area:

• bun run test:provider
• bun run test:provider-recommendation
• bun test path/to/file.test.ts

Recommended contributor validation before opening a PR:

• bun run build
• bun run smoke
• bun run test:coverage for broader unit coverage when your change affects shared runtime or provider logic
• focused bun test ... runs for the files and flows you changed

Coverage output is written to coverage/lcov.info, and Zero CLI also generates a git-activity-style heatmap at coverage/index.html.

Repository Structure

• src/ - core CLI/runtime
• scripts/ - build, verification, and maintenance scripts
• docs/ - setup, contributor, and project documentation
• python/ - standalone Python helpers and their tests
• vscode-extension/zero-vscode/ - VS Code extension
• .github/ - repo automation, templates, and CI configuration
• bin/ - CLI launcher entrypoints

VS Code Extension

The repo includes a VS Code extension in vscode-extension/zero-vscode for Zero CLI launch integration, provider-aware control-center UI, and theme support.

Security

If you believe you found a security issue, see SECURITY.md.

Community

• Use GitHub Discussions for Q&A, ideas, and community conversation
• Use GitHub Issues for confirmed bugs and actionable feature work

Contributing

Contributions are welcome.

For larger changes, open an issue first so the scope is clear before implementation. Helpful validation commands include:

• bun run build
• bun run test:coverage
• bun run smoke
• focused bun test ... runs for files and flows you changed

Disclaimer

Zero CLI is an independent community project and is not affiliated with, endorsed by, or sponsored by Anthropic.

Zero CLI originated from the Claude Code codebase and has since been substantially modified to support multiple providers and open use. "Claude" and "Claude Code" are trademarks of Anthropic PBC. See LICENSE for details.

License

See LICENSE.