### Start OpenFang daemon examples

Source: https://github.com/rightnow-ai/openfang/blob/main/docs/cli-reference.md

Commands for starting the daemon with default or custom configurations.

```bash
# Start with default config
openfang start

# Start with custom config
openfang start --config /path/to/config.toml
```

--------------------------------

### Install and Start OpenFang (Linux/macOS)

Source: https://github.com/rightnow-ai/openfang/blob/main/README.md

Use this command to install OpenFang via curl and then initialize and start the agent system. The dashboard will be accessible at http://localhost:4200.

```bash
curl -fsSL https://openfang.sh/install | sh
openfang init
openfang start
# Dashboard live at http://localhost:4200
```

--------------------------------

### Install and Run vLLM Server

Source: https://github.com/rightnow-ai/openfang/blob/main/docs/providers.md

Instructions for installing vLLM and starting its OpenAI-compatible API server. No API key is required for local use.

```bash
pip install vllm
```

```bash
python -m vllm.entrypoints.openai.api_server --model <model-name>
```

--------------------------------

### OpenFang First-time Setup

Source: https://github.com/rightnow-ai/openfang/blob/main/docs/cli-reference.md

Steps to set up OpenFang for the first time, including exporting API keys, initializing the application, and starting the daemon.

```bash
# 1. Set your API key
export GROQ_API_KEY="gsk_your_key_here"

# 2. Initialize OpenFang
openfang init --quick

# 3. Start the daemon
openfang start
```

--------------------------------

### Install and Start OpenFang (Windows)

Source: https://github.com/rightnow-ai/openfang/blob/main/README.md

Use this PowerShell command to install OpenFang and then initialize and start the agent system. The dashboard will be accessible at http://localhost:4200.

```powershell
irm https://openfang.sh/install.ps1 | iex
openfang init
openfang start
```

--------------------------------

### Initialize OpenFang

Source: https://github.com/rightnow-ai/openfang/blob/main/README.md

Run the openfang init command to start the initialization process, which includes provider setup.

```bash
# 2. Initialize. Walks you through provider setup.
openfang init
```

--------------------------------

### Initialize and Start OpenFang

Source: https://github.com/rightnow-ai/openfang/blob/main/docs/README.md

Quick start command sequence to initialize the environment and launch the OpenFang daemon.

```bash
export GROQ_API_KEY="your-key"
openfang init && openfang start
# Open http://127.0.0.1:4200
```

--------------------------------

### OpenFang daemon output example

Source: https://github.com/rightnow-ai/openfang/blob/main/docs/cli-reference.md

Example console output upon successful daemon startup.

```text
  OpenFang Agent OS v0.1.0

  Starting daemon...

  [ok] Kernel booted (groq/llama-3.3-70b-versatile)
  [ok] 50 models available
  [ok] 3 agent(s) loaded

  API:        http://127.0.0.1:4200
  Dashboard:  http://127.0.0.1:4200/
  Provider:   groq
  Model:      llama-3.3-70b-versatile

  hint: Open the dashboard in your browser, or run `openfang chat`
  hint: Press Ctrl+C to stop the daemon
```

--------------------------------

### Install OpenFang from source

Source: https://github.com/rightnow-ai/openfang/blob/main/docs/cli-reference.md

Installs the CLI tool using Cargo.

```bash
cargo install --path crates/openfang-cli
```

--------------------------------

### Manual PowerShell Setup for OpenFang

Source: https://github.com/rightnow-ai/openfang/blob/main/docs/VERTEX_AI_LOCAL_TESTING.md

Manually set up environment variables and start OpenFang using PowerShell commands. Ensure to clear proxy settings and pre-fetch the OAuth token for Windows.

```powershell
# 1. Kill any existing instances
taskkill /F /IM openfang.exe 2>$null

# 2. Set environment variables (CRITICAL: clear proxy!)
$env:HTTPS_PROXY = ""
$env:HTTP_PROXY = ""
$env:GOOGLE_APPLICATION_CREDENTIALS = "C:\Users\at384\Downloads\osc\dbg-grcit-dev-e1-c79e5571a5a7.json"

# 3. Pre-fetch OAuth token (IMPORTANT: avoids subprocess issues on Windows)
$env:VERTEX_AI_ACCESS_TOKEN = gcloud auth print-access-token

# 4. Start OpenFang
cd C:\Users\at384\Downloads\osc\dllm\openfang
.\target\debug\openfang.exe start
```

--------------------------------

### Install a Skill

Source: https://github.com/rightnow-ai/openfang/blob/main/docs/cli-reference.md

Install a skill from various sources including a local directory, a Git URL, or the FangHub marketplace. The command handles verification and scanning for remote installations.

```bash
openfang skill install <SOURCE>
```

```bash
# Install from local directory
openfang skill install ./my-skill/
```

```bash
# Install from FangHub
openfang skill install web-search
```

```bash
# Install an OpenClaw-format skill
openfang skill install ./openclaw-skill/
```

--------------------------------

### Build and Install OpenFang CLI from Source

Source: https://github.com/rightnow-ai/openfang/blob/main/docs/getting-started.md

Clones the OpenFang repository and installs the CLI from source using Cargo. This is an alternative to installing directly from the Git repository.

```bash
git clone https://github.com/RightNow-AI/openfang.git
cd openfang
cargo install --path crates/openfang-cli
```

--------------------------------

### GET /skill/list

Source: https://github.com/rightnow-ai/openfang/blob/main/docs/cli-reference.md

List all installed skills.

```APIDOC
## GET /skill/list

### Description
List all installed skills.
```

--------------------------------

### Install Manual Skill Directory

Source: https://github.com/rightnow-ai/openfang/blob/main/docs/skill-development.md

Install a skill directory after manually creating the required manifest.

```bash
openfang skill install /path/to/skill-directory
```

--------------------------------

### IDE Integration Examples

Source: https://github.com/rightnow-ai/openfang/blob/main/docs/mcp-a2a.md

Configuration examples for integrating OpenFang MCP servers with IDEs like Cursor and Claude Desktop.

```APIDOC
## IDE Integration Examples

### Description
Provides configuration snippets for integrating OpenFang MCP servers into popular IDEs that support the MCP protocol.

### Method
N/A (Configuration)

### Endpoint
N/A

### Parameters
N/A

### Request Example
#### Cursor / VS Code (MCP extension)
Add the following to your MCP configuration file (e.g., `.cursor/mcp.json` or VS Code MCP settings):

```json
{
  "mcpServers": {
    "openfang": {
      "command": "openfang",
      "args": ["mcp"]
    }
  }
}
```

#### Claude Desktop
Add the following to `claude_desktop_config.json`:

```json
{
  "mcpServers": {
    "openfang": {
      "command": "openfang",
      "args": ["mcp"],
      "env": {}
    }
  }
}
```

### Response
N/A
```

--------------------------------

### Spawn OpenFang Agent using Built-in Template

Source: https://github.com/rightnow-ai/openfang/blob/main/docs/getting-started.md

Spawns the 'hello-world' agent using its predefined manifest file. This is a quick way to get started with a basic agent.

```bash
openfang agent spawn agents/hello-world/agent.toml
```

--------------------------------

### Interactive Telegram Channel Setup

Source: https://github.com/rightnow-ai/openfang/blob/main/docs/channel-adapters.md

Use the `openfang channel setup telegram` command for an interactive walkthrough of setting up the Telegram adapter. This command guides you through the necessary steps.

```bash
openfang channel setup telegram
```

--------------------------------

### Install OpenFang via shell script

Source: https://github.com/rightnow-ai/openfang/blob/main/docs/cli-reference.md

Downloads and runs the automated installation script.

```bash
curl -fsSL https://get.openfang.ai | sh
```

--------------------------------

### Start OpenFang Channels with Debug Logging

Source: https://github.com/rightnow-ai/openfang/blob/main/docs/troubleshooting.md

Run OpenFang with RUST_LOG set to debug for the channels crate to get detailed logs for troubleshooting channel adapter issues.

```bash
RUST_LOG=openfang_channels=debug openfang start
```

--------------------------------

### Install Skill Request Body

Source: https://github.com/rightnow-ai/openfang/blob/main/docs/api-reference.md

Request body for installing a skill. Specify the source path or URL and whether to verify the installation.

```json
{
  "source": "/path/to/skill",
  "verify": true
}
```

--------------------------------

### Docker Compose Configuration for OpenFang

Source: https://github.com/rightnow-ai/openfang/blob/main/docs/configuration.md

Example Docker Compose setup for OpenFang, including port mapping, volume mounting, and environment variables for binding and API keys.

```yaml
services:
  openfang:
    image: ghcr.io/rightnow-ai/openfang:latest
    ports:
      - "4200:4200"
    volumes:
      - openfang-data:/data
    environment:
      - OPENFANG_LISTEN=0.0.0.0:4200          # required: bind to all interfaces inside the container
      - OPENFANG_API_KEY=${OPENFANG_API_KEY}  # strongly recommended when exposing
      - ANTHROPIC_API_KEY=${ANTHROPIC_API_KEY:-}
volumes:
  openfang-data:
```

--------------------------------

### Start Daemon with API Keys

Source: https://github.com/rightnow-ai/openfang/blob/main/CLAUDE.md

Start the OpenFang daemon with the necessary API keys. Ensure you wait for the daemon to fully boot before proceeding.

```bash
GROQ_API_KEY=<key> target/release/openfang.exe start &
sleep 6  # Wait for full boot
curl -s http://127.0.0.1:4200/api/health  # Verify it's up
```

--------------------------------

### Install Skill Response

Source: https://github.com/rightnow-ai/openfang/blob/main/docs/api-reference.md

Response indicating successful skill installation, including the skill name and version.

```json
{
  "status": "installed",
  "skill": "my-custom-skill",
  "version": "1.0.0"
}
```

--------------------------------

### Linux/macOS Install Script Execution

Source: https://github.com/rightnow-ai/openfang/blob/main/docs/production-checklist.md

Download and execute the install script for Linux and macOS using curl. Verify the installation by checking the version.

```bash
curl -sSf https://openfang.sh | sh
openfang --version  # Should print v0.1.0
```

--------------------------------

### Start vLLM OpenAI API Server

Source: https://github.com/rightnow-ai/openfang/blob/main/docs/troubleshooting.md

Command to start a vLLM server compatible with the OpenAI API.

```bash
# vLLM
python -m vllm.entrypoints.openai.api_server --model ...
```

--------------------------------

### Install OpenFang (macOS/Linux)

Source: https://github.com/rightnow-ai/openfang/blob/main/README.md

Download and execute the OpenFang installation script using curl.

```bash
# 1. Install (macOS/Linux)
curl -fsSL https://openfang.sh/install | sh
```

--------------------------------

### POST /api/skills/install

Source: https://github.com/rightnow-ai/openfang/blob/main/docs/api-reference.md

Installs a skill from a local path or URL.

```APIDOC
## POST /api/skills/install

### Description
Install a skill from a local path or URL. The skill manifest is verified (SHA256 checksum) and scanned for prompt injection before installation.

### Method
POST

### Endpoint
/api/skills/install

### Parameters
#### Request Body
- **source** (string) - Required - Path or URL to the skill
- **verify** (boolean) - Required - Whether to verify the skill

### Response
#### Success Response (201)
- **status** (string) - Status of the operation
- **skill** (string) - Name of the installed skill
- **version** (string) - Version of the installed skill

#### Response Example
{
  "status": "installed",
  "skill": "my-custom-skill",
  "version": "1.0.0"
}
```

--------------------------------

### POST /skill/install

Source: https://github.com/rightnow-ai/openfang/blob/main/docs/cli-reference.md

Install a skill from a source.

```APIDOC
## POST /skill/install

### Description
Install a skill from a local directory, git URL, or FangHub marketplace.

### Endpoint
openfang skill install <SOURCE>

### Parameters
#### Path Parameters
- **SOURCE** (string) - Required - Skill name, local directory path, or git URL.
```

--------------------------------

### Install WhatsApp Gateway Dependencies

Source: https://github.com/rightnow-ai/openfang/blob/main/README.md

Navigate to the whatsapp-gateway package directory and install its dependencies using npm.

```bash
cd packages/whatsapp-gateway
npm install
```

--------------------------------

### Workflow Step Configuration Example

Source: https://github.com/rightnow-ai/openfang/blob/main/docs/workflows.md

An example of a single workflow step configuration, specifying agent, prompt, mode, and error handling.

```json
{
  "name": "research",
  "agent_name": "web-search",
  "prompt": "Search for the latest news on {{topic}}",
  "mode": "sequential",
  "timeout_secs": 60,
  "error_mode": "retry",
  "max_retries": 3,
  "output_var": "search_results"
}
```

--------------------------------

### POST /api/clawhub/install

Source: https://github.com/rightnow-ai/openfang/blob/main/docs/api-reference.md

Installs a skill from ClawHub.

```APIDOC
## POST /api/clawhub/install

### Description
Install a skill from ClawHub. Downloads, verifies SHA256 checksum, scans for prompt injection, and converts SKILL.md format to OpenFang skill.toml automatically.

### Method
POST

### Endpoint
/api/clawhub/install

### Parameters
#### Request Body
- **slug** (string) - Required - The skill slug to install

### Response
#### Success Response (201)
- **status** (string) - Status of the operation
- **skill** (string) - Name of the installed skill
- **version** (string) - Version of the installed skill
- **converted_from** (string) - Source format

#### Response Example
{
  "status": "installed",
  "skill": "data-pipeline",
  "version": "1.2.0",
  "converted_from": "SKILL.md"
}
```

--------------------------------

### Install OpenFang using Raw GitHub Redirect

Source: https://github.com/rightnow-ai/openfang/blob/main/docs/production-checklist.md

Provides a temporary method to install OpenFang using a direct curl command pointing to the raw install script on GitHub. This is a fallback until the `openfang.sh` domain is properly configured.

```bash
curl -sSf https://raw.githubusercontent.com/RightNow-AI/openfang/main/scripts/install.sh | sh
```

--------------------------------

### Start OpenFang Daemon

Source: https://github.com/rightnow-ai/openfang/blob/main/README.md

Start the OpenFang daemon to make the dashboard and API available. The dashboard is accessible at http://localhost:4200.

```bash
# 3. Start the daemon
openfang start
```

--------------------------------

### Install OpenFang CLI via Shell

Source: https://github.com/rightnow-ai/openfang/blob/main/docs/getting-started.md

Downloads and installs the latest OpenFang CLI binary to `~/.openfang/bin/`. This method is suitable for Linux and macOS.

```bash
curl -sSf https://openfang.sh | sh
```

--------------------------------

### Start MCP Server

Source: https://github.com/rightnow-ai/openfang/blob/main/docs/getting-started.md

Command to start the MCP server, typically used for standard input/output communication.

```bash
openfang mcp                           # Start MCP server (stdio)
```

--------------------------------

### OpenFang Configuration Example

Source: https://github.com/rightnow-ai/openfang/blob/main/docs/VERTEX_AI_LOCAL_TESTING.md

Example TOML configuration file for OpenFang, specifying the default model provider, memory decay rate, and network listen address.

```toml
[default_model]
provider = "vertex-ai"
model = "gemini-2.0-flash"

[memory]
decay_rate = 0.05

[network]
listen_addr = "127.0.0.1:4200"
```

--------------------------------

### List Installed Skills

Source: https://github.com/rightnow-ai/openfang/blob/main/docs/cli-reference.md

This command lists all skills that are currently installed. It shows the skill's name, version, associated tools, and a brief description. Skills are loaded from `~/.openfang/skills/` and bundled ones.

```bash
openfang skill list
```

--------------------------------

### List Installed Skills

Source: https://github.com/rightnow-ai/openfang/blob/main/docs/api-reference.md

Retrieve a list of all installed skills, including their name, version, runtime, description, and bundled status.

```json
{
  "skills": [
    {
      "name": "github",
      "version": "1.0.0",
      "runtime": "prompt_only",
      "description": "GitHub integration for issues, PRs, and repos",
      "bundled": true
    },
    {
      "name": "docker",
      "version": "1.0.0",
      "runtime": "prompt_only",
      "description": "Docker container management",
      "bundled": true
    }
  ],
  "total": 60
}
```

--------------------------------

### Knowledge Graph Query Examples

Source: https://github.com/rightnow-ai/openfang/blob/main/crates/openfang-hands/bundled/infisical-sync/SKILL.md

Examples of queries to the knowledge graph for retrieving information about secrets, projects, and their relationships.

```shell
knowledge_query type=secret                  // list all known secrets
```

```shell
knowledge_query type=secret_project          // list all projects
```

```shell
knowledge_query relation=belongs_to target=<project_id>  // secrets in a project
```

--------------------------------

### Content Pillars Example

Source: https://github.com/rightnow-ai/openfang/blob/main/crates/openfang-hands/bundled/twitter/SKILL.md

An example of defining 3-5 core topics for content strategy, with a suggested percentage allocation for a tech founder.

```text
Example for a tech founder:
  Pillar 1: AI & Machine Learning (40% of content)
  Pillar 2: Startup Building (30% of content)
  Pillar 3: Engineering Culture (20% of content)
  Pillar 4: Personal Growth (10% of content)
```

--------------------------------

### Initialize OpenFang examples

Source: https://github.com/rightnow-ai/openfang/blob/main/docs/cli-reference.md

Demonstrates interactive and non-interactive initialization workflows.

```bash
# Interactive setup
openfang init

# Non-interactive (CI/scripts)
export GROQ_API_KEY="gsk_..."
openfang init --quick
```

--------------------------------

### Install OpenFang CLI via PowerShell

Source: https://github.com/rightnow-ai/openfang/blob/main/docs/getting-started.md

Downloads, verifies the checksum, and installs the latest OpenFang CLI binary to your user PATH. This method is for Windows.

```powershell
irm https://openfang.sh/install.ps1 | iex
```

--------------------------------

### Test Linux/macOS Install Script

Source: https://github.com/rightnow-ai/openfang/blob/main/docs/production-checklist.md

Executes the `install.sh` script locally to test the installation process on Linux or macOS. This should be tested against a real GitHub release after the first tag is created.

```bash
bash scripts/install.sh
```

--------------------------------

### GET /api/skills

Source: https://github.com/rightnow-ai/openfang/blob/main/docs/api-reference.md

Lists all installed skills.

```APIDOC
## GET /api/skills

### Description
List all installed skills.

### Method
GET

### Endpoint
/api/skills

### Response
#### Success Response (200)
- **skills** (array) - List of installed skills
- **total** (integer) - Total count of skills

#### Response Example
{
  "skills": [
    {
      "name": "github",
      "version": "1.0.0",
      "runtime": "prompt_only",
      "description": "GitHub integration for issues, PRs, and repos",
      "bundled": true
    }
  ],
  "total": 60
}
```

--------------------------------

### Clone and build the repository

Source: https://github.com/rightnow-ai/openfang/blob/main/CONTRIBUTING.md

Initial setup commands to download the source code and perform the first build.

```bash
git clone https://github.com/RightNow-AI/openfang.git
cd openfang
cargo build
```

--------------------------------

### Start OpenFang daemon

Source: https://github.com/rightnow-ai/openfang/blob/main/docs/cli-reference.md

Boots the kernel and API server.

```bash
openfang start [--config <PATH>]
```

--------------------------------

### Setup channel integration

Source: https://github.com/rightnow-ai/openfang/blob/main/docs/cli-reference.md

Launches an interactive wizard to configure a specific channel or all channels if no argument is provided.

```bash
openfang channel setup [<CHANNEL>]
```

```bash
# Interactive picker
openfang channel setup

# Direct setup
openfang channel setup telegram
openfang channel setup discord
openfang channel setup slack
```

--------------------------------

### Reset OpenFang Installation

Source: https://github.com/rightnow-ai/openfang/blob/main/docs/troubleshooting.md

Use this command to completely remove the OpenFang configuration and data, allowing for a fresh start.

```bash
rm -rf ~/.openfang
openfang init  # Start fresh
```

--------------------------------

### Initialize OpenFang workspace

Source: https://github.com/rightnow-ai/openfang/blob/main/docs/cli-reference.md

Sets up the configuration and directory structure.

```bash
openfang init [--quick]
```

--------------------------------

### Retrieve MCP Server Status

Source: https://github.com/rightnow-ai/openfang/blob/main/docs/mcp-a2a.md

Example JSON response from the GET /api/mcp/servers endpoint showing configured and connected servers.

```json
{
  "configured": [
    {
      "name": "github",
      "transport": { "type": "stdio", "command": "npx", "args": [...] },
      "timeout_secs": 30,
      "env": ["GITHUB_PERSONAL_ACCESS_TOKEN"]
    }
  ],
  "connected": [
    {
      "name": "github",
      "tools_count": 12,
      "tools": [
        { "name": "mcp_github_create_issue", "description": "[MCP:github] Create a GitHub issue" },
        { "name": "mcp_github_search_repos", "description": "[MCP:github] Search repositories" }
      ],
      "connected": true
    }
  ]
}
```

--------------------------------

### Get Cron Job Status Response

Source: https://github.com/rightnow-ai/openfang/blob/main/docs/api-reference.md

Example JSON response for retrieving the status of a specific cron job, including run history and configuration.

```json
{
  "job": {
    "id": "550e8400-e29b-41d4-a716-446655440000",
    "agent_id": "7c9e6679-7425-40de-944b-e07fc1f90ae7",
    "name": "daily-report",
    "enabled": true,
    "schedule": { "kind": "every", "every_secs": 3600 },
    "action": {
      "kind": "agent_turn",
      "message": "Generate the daily report",
      "timeout_secs": 120
    },
    "delivery": { "kind": "none" },
    "created_at": "2026-03-15T10:30:00Z",
    "last_run": "2026-03-16T09:00:00Z",
    "next_run": "2026-03-16T10:00:00Z"
  },
  "one_shot": false,
  "last_status": "ok",
  "consecutive_errors": 0
}
```

--------------------------------

### Build Production Binaries

Source: https://github.com/rightnow-ai/openfang/blob/main/docs/desktop.md

Commands to generate platform-specific installers.

```bash
cd crates/openfang-desktop
cargo tauri build
```

--------------------------------

### Dockerfile Best Practices

Source: https://github.com/rightnow-ai/openfang/blob/main/crates/openfang-skills/bundled/docker/SKILL.md

Examples of recommended Dockerfile instructions for image optimization and security.

```dockerfile
USER
```

```dockerfile
COPY --from=builder
```

```dockerfile
HEALTHCHECK
```

```dockerfile
COPY
```

```dockerfile
ADD
```

--------------------------------

### Full OpenFang Agent OS Configuration Example

Source: https://github.com/rightnow-ai/openfang/blob/main/docs/configuration.md

A comprehensive TOML configuration file demonstrating all major sections and options for the OpenFang Agent OS. This includes top-level settings, default and fallback LLM providers, memory configuration, and network settings.

```toml
# ============================================================
# OpenFang Agent OS -- Complete Configuration Reference
# ============================================================

# --- Top-level fields ---
home_dir = "~/.openfang"             # OpenFang home directory
data_dir = "~/.openfang/data"        # SQLite databases and data files
log_level = "info"                   # trace | debug | info | warn | error
api_listen = "127.0.0.1:50051"      # HTTP/WS API bind address
network_enabled = false              # Enable OFP peer-to-peer network
api_key = ""                         # API Bearer token (empty = unauthenticated)
mode = "default"                     # stable | default | dev
language = "en"                      # Locale for CLI/messages
usage_footer = "full"                # off | tokens | cost | full

# --- Default LLM Provider ---
[default_model]
provider = "anthropic"
model = "claude-sonnet-4-20250514"
api_key_env = "ANTHROPIC_API_KEY"
# base_url = "https://api.anthropic.com"  # Optional override

# --- Fallback Providers ---
[[fallback_providers]]
provider = "ollama"
model = "llama3.2:latest"
api_key_env = ""
# base_url = "http://localhost:11434"  # Uses catalog default if omitted

[[fallback_providers]]
provider = "groq"
model = "llama-3.3-70b-versatile"
api_key_env = "GROQ_API_KEY"

# --- Memory ---
[memory]
# sqlite_path = "~/.openfang/data/openfang.db"  # Auto-resolved if omitted
embedding_model = "all-MiniLM-L6-v2"
consolidation_threshold = 10000
decay_rate = 0.1

# --- Network (OFP Wire Protocol) ---
[network]
listen_addresses = ["/ip4/0.0.0.0/tcp/0"]
bootstrap_peers = []
mdns_enabled = true
max_peers = 50
shared_secret = ""                   # Required when network_enabled = true

```

--------------------------------

### Windows PowerShell Install Script Execution

Source: https://github.com/rightnow-ai/openfang/blob/main/docs/production-checklist.md

Download and execute the install script for Windows using Invoke-RestMethod (irm). Verify the installation by checking the version.

```powershell
irm https://openfang.sh/install.ps1 | iex
openfang --version
```

--------------------------------

### Initialize OpenFang Configuration

Source: https://github.com/rightnow-ai/openfang/blob/main/docs/getting-started.md

Creates the `~/.openfang/` directory and a default `config.toml` file. This is the first step in configuring OpenFang.

```bash
openfang init
```

--------------------------------

### Install OpenFang Skills

Source: https://github.com/rightnow-ai/openfang/blob/main/MIGRATION.md

After migration, skills from OpenClaw are not automatically converted. Use this command to install skills into OpenFang, which will automatically detect and convert OpenClaw-format skills during installation.

```bash
openfang skill install <skill-name-or-path>
```

--------------------------------

### Groq Provider Setup

Source: https://github.com/rightnow-ai/openfang/blob/main/docs/providers.md

Set the GROQ_API_KEY environment variable after creating an API key on the Groq console. The free tier is rate-limited but usable.

```bash
export GROQ_API_KEY="gsk_..."
```

--------------------------------

### GET /a2a/tasks/{id}

Source: https://github.com/rightnow-ai/openfang/blob/main/docs/api-reference.md

Get the status and result of an A2A task.

```APIDOC
## GET /a2a/tasks/{id}

### Description
Get the status and result of an A2A task.

### Method
GET

### Endpoint
/a2a/tasks/{id}

### Parameters
#### Path Parameters
- **id** (string) - Required - The task ID

### Response
#### Success Response (200)
- **task_id** (string) - Task identifier
- **status** (string) - Current status
- **result** (object) - Task result
```

--------------------------------

### GET /api/models/{id}

Source: https://github.com/rightnow-ai/openfang/blob/main/docs/api-reference.md

Get detailed information about a specific model.

```APIDOC
## GET /api/models/{id}

### Description
Get detailed information about a specific model.

### Method
GET

### Endpoint
/api/models/{id}

### Parameters
#### Path Parameters
- **id** (string) - Required - The model ID

### Response
#### Success Response (200)
- **id** (string) - Model ID
- **provider** (string) - Provider name
- **display_name** (string) - Human readable name
- **tier** (string) - Model tier
- **context_window** (integer) - Context window size
- **input_cost_per_1m** (float) - Input cost
- **output_cost_per_1m** (float) - Output cost
- **supports_tools** (boolean) - Tool support flag
- **supports_vision** (boolean) - Vision support flag
- **supports_streaming** (boolean) - Streaming support flag

#### Response Example
{
  "id": "llama-3.3-70b-versatile",
  "provider": "groq",
  "display_name": "Llama 3.3 70B",
  "tier": "fast",
  "context_window": 131072,
  "input_cost_per_1m": 0.59,
  "output_cost_per_1m": 0.79,
  "supports_tools": true,
  "supports_vision": false,
  "supports_streaming": true
}
```

--------------------------------

### Initialize OpenFang Workspace

Source: https://github.com/rightnow-ai/openfang/blob/main/MIGRATION.md

Run this command to initialize a new OpenFang workspace, creating the `~/.openfang/` directory with a default `config.toml` file. This is the first step in a manual migration.

```bash
openfang init
```

--------------------------------

### GET /api/clawhub/skill/{slug}

Source: https://github.com/rightnow-ai/openfang/blob/main/docs/api-reference.md

Gets detailed information about a specific ClawHub skill.

```APIDOC
## GET /api/clawhub/skill/{slug}

### Description
Get detailed information about a specific ClawHub skill.

### Method
GET

### Endpoint
/api/clawhub/skill/{slug}

### Parameters
#### Path Parameters
- **slug** (string) - Required - The skill slug

### Response
#### Success Response (200)
- **slug** (string) - Skill slug
- **name** (string) - Skill name
- **description** (string) - Skill description
- **author** (string) - Skill author
- **version** (string) - Skill version
- **runtime** (string) - Runtime environment
- **readme** (string) - Skill documentation
- **sha256** (string) - SHA256 checksum

#### Response Example
{
  "slug": "data-pipeline",
  "name": "Data Pipeline",
  "description": "ETL data pipeline automation",
  "author": "clawhub-community",
  "version": "1.2.0",
  "runtime": "python",
  "readme": "# Data Pipeline\n\nAutomated ETL...",
  "sha256": "a1b2c3d4..."
}
```

--------------------------------

### Verify OpenFang Installation

Source: https://github.com/rightnow-ai/openfang/blob/main/docs/getting-started.md

Checks if the OpenFang CLI is installed correctly by displaying its version.

```bash
openfang --version
```

--------------------------------

### OpenFang Channel Setup and Management

Source: https://github.com/rightnow-ai/openfang/blob/main/docs/cli-reference.md

Commands for setting up, testing, enabling, and disabling various communication channels.

```bash
# Interactive channel picker
openfang channel setup

# Direct channel setup
openfang channel setup telegram

# Check channel status
openfang channel list

# Test a channel
openfang channel test telegram

# Enable/disable channels
openfang channel enable discord
openfang channel disable slack
```

--------------------------------

### Initialize OpenFang

Source: https://github.com/rightnow-ai/openfang/blob/main/docs/getting-started.md

Use this command to initialize the OpenFang configuration directory.

```bash
openfang init                          # Initialize ~/.openfang/
```

--------------------------------

### Install OpenClaw Skill

Source: https://github.com/rightnow-ai/openfang/blob/main/docs/skill-development.md

Use the CLI to install an OpenClaw-format skill, which triggers automatic detection and conversion.

```bash
openfang skill install /path/to/openclaw-skill
```

--------------------------------

### Test GET Endpoints

Source: https://github.com/rightnow-ai/openfang/blob/main/CLAUDE.md

Verify that GET endpoints return real data, not empty or null values.

```bash
curl -s http://127.0.0.1:4200/api/<new-endpoint>
```

--------------------------------

### Manage Skills

Source: https://github.com/rightnow-ai/openfang/blob/main/docs/getting-started.md

Commands for installing, listing, searching, and creating skills.

```bash
openfang skill install <source>        # Install a skill
```

```bash
openfang skill list                    # List installed skills
```

```bash
openfang skill search <query>          # Search FangHub
```

```bash
openfang skill create                  # Scaffold a new skill
```

--------------------------------

### Build OpenFang from workspace

Source: https://github.com/rightnow-ai/openfang/blob/main/docs/cli-reference.md

Compiles the CLI binary from the local workspace.

```bash
cargo build --release -p openfang-cli
# Binary: target/release/openfang (or openfang.exe on Windows)
```

--------------------------------

### Update Rust Toolchain and Install OpenFang

Source: https://github.com/rightnow-ai/openfang/blob/main/docs/troubleshooting.md

If `cargo install` fails due to compilation errors, update your stable Rust toolchain to version 1.75 or later. Ensure you have necessary system dependencies like `pkg-config`, `libssl-dev`, and `libsqlite3-dev` installed.

```bash
rustup update stable
rustup default stable
rustc --version  # Need 1.75+
```

```bash
# Debian/Ubuntu
sudo apt install pkg-config libssl-dev libsqlite3-dev

# Fedora
sudo dnf install openssl-devel sqlite-devel
```

--------------------------------

### Quick Setup: Set LLM Provider API Key

Source: https://github.com/rightnow-ai/openfang/blob/main/docs/providers.md

Set the environment variable for your chosen LLM provider to authenticate it. Local providers do not require a key. For Gemini, either GEMINI_API_KEY or GOOGLE_API_KEY can be used.

```bash
export GEMINI_API_KEY="your-key"        # Free tier available
# OR
export GROQ_API_KEY="your-key"          # Free tier available
# OR
export ANTHROPIC_API_KEY="your-key"
# OR
export OPENAI_API_KEY="your-key"
```

--------------------------------

### Start a chat session

Source: https://github.com/rightnow-ai/openfang/blob/main/docs/cli-reference.md

Initiates a chat with an agent, either via a running daemon or by booting an in-process kernel.

```bash
# Chat with the default agent
openfang chat

# Chat with a specific agent by name
openfang chat coder

# Chat with a specific agent by UUID
openfang chat a1b2c3d4-e5f6-7890-abcd-ef1234567890
```

--------------------------------

### Update OpenFang via Install Script

Source: https://github.com/rightnow-ai/openfang/blob/main/docs/troubleshooting.md

Re-run the official install script to update OpenFang to the latest release.

```bash
curl -fsSL https://openfang.sh/install | sh
```

--------------------------------

### Remove an Installed Skill

Source: https://github.com/rightnow-ai/openfang/blob/main/docs/cli-reference.md

Uninstall a skill by specifying its name. This command removes the skill from your local OpenFang installation.

```bash
openfang skill remove <NAME>
```

```bash
openfang skill remove web-search
```

--------------------------------

### Python Skill Example: Web Summarizer

Source: https://github.com/rightnow-ai/openfang/blob/main/docs/skill-development.md

An example of a Python skill that can summarize a given URL or extract links from it.

```APIDOC
## Example: Web Summarizer

`src/main.py`:

```python
#!/usr/bin/env python3
"""OpenFang skill: web-summarizer"""
import json
import sys
import urllib.request


def summarize_url(url: str) -> str:
    """Fetch a URL and return a basic summary."""
    req = urllib.request.Request(url, headers={"User-Agent": "OpenFang-Skill/1.0"})
    with urllib.request.urlopen(req, timeout=30) as resp:
        content = resp.read().decode("utf-8", errors="replace")

    # Simple extraction: first 500 chars as summary
    text = content[:500].strip()
    return f"Summary of {url}:\n{text}..."


def extract_links(url: str) -> str:
    """Extract all links from a web page."""
    import re

    req = urllib.request.Request(url, headers={"User-Agent": "OpenFang-Skill/1.0"})
    with urllib.request.urlopen(req, timeout=30) as resp:
        content = resp.read().decode("utf-8", errors="replace")

    links = re.findall(r'href="(https?://[^\"]+)"', content)
    unique_links = list(dict.fromkeys(links))
    return "\n".join(unique_links[:50])


def main():
    payload = json.loads(sys.stdin.read())
    tool_name = payload["tool"]
    input_data = payload["input"]

    try:
        if tool_name == "summarize_url":
            result = summarize_url(input_data["url"])
        elif tool_name == "extract_links":
            result = extract_links(input_data["url"])
        else:
            print(json.dumps({"error": f"Unknown tool: {tool_name}"}))
            return

        print(json.dumps({"result": result}))
    except Exception as e:
        print(json.dumps({"error": str(e)}))


if __name__ == "__main__":
    main()
```
```

--------------------------------

### Alpaca API Authentication Setup

Source: https://github.com/rightnow-ai/openfang/blob/main/crates/openfang-hands/bundled/trader/SKILL.md

Bash script to set up environment variables for Alpaca API base URLs (paper and live trading) and authentication headers.

```bash
# Paper trading (ALWAYS start here)
BASE_URL="https://paper-api.alpaca.markets"

# Live trading (only after paper validation)
# BASE_URL="https://api.alpaca.markets"

# Data API (same for both paper and live)
DATA_URL="https://data.alpaca.markets"

# Auth headers (required on every request)
HEADERS="-H 'APCA-API-KEY-ID: $ALPACA_API_KEY' -H 'APCA-API-SECRET-KEY: $ALPACA_SECRET_KEY'"
```