synthflow-voice-ai

Synthflow Skills

Agent skills for Synthflow — the platform for building AI voice agents. These skills follow the Agent Skills specification and can be used with any compatible AI coding assistant including Claude Code, Cursor, VS Code Copilot, Gemini CLI, and more.

Installation

Option 1: npx skills (works with all agents)

npx skills add SynthFlowAI/synthflow-skills

Install specific skills:

npx skills add SynthFlowAI/synthflow-skills --skill create-assistant
npx skills add SynthFlowAI/synthflow-skills --skill create-call

Install for a specific agent:

npx skills add SynthFlowAI/synthflow-skills -a claude-code
npx skills add SynthFlowAI/synthflow-skills -a cursor

Option 2: Claude Code Plugin (native integration)

/plugin marketplace add SynthFlowAI/synthflow-skills
/plugin install synthflow-voice-ai@synthflow-skills

Option 3: Manual installation

Copy any skill directory into your project's .claude/skills/ (for Claude Code), .cursor/skills/ (for Cursor), or the equivalent skills directory for your agent.

Supported agents

Agent	Config File	Auto-detected
Claude Code	.mcp.json	Yes
Cursor	.cursor/mcp.json	Yes
VS Code Copilot	.vscode/mcp.json	Yes

Requires: Node.js (for npx). Uses mcp-remote to bridge the remote server. No API key needed for the docs server.

Manual setup

If your agent doesn't auto-detect MCP configs:

Claude Code:

claude mcp add synthflow-docs -- npx -y mcp-remote https://docs.synthflow.ai/_mcp/server

Any agent (JSON config):

{
  "mcpServers": {
    "synthflow-docs": {
      "command": "npx",
      "args": ["-y", "mcp-remote", "https://docs.synthflow.ai/_mcp/server"]
    }
  }
}

Available Skills

Skill	Description
setup-api-key	Guide through obtaining and configuring a Synthflow API key
create-assistant	Create AI voice assistants with LLMs, voices, prompts, and advanced settings
create-call	Initiate outbound phone calls through AI agents
manage-actions	Create, update, and attach actions — transfers, SMS, webhooks, booking, extraction, evals
create-simulation	Create and run simulation test suites to evaluate agent behavior
create-eval	Define custom evaluations to score call quality and outcomes

Configuration

All skills require a Synthflow API key. Set it as an environment variable:

export SYNTHFLOW_API_KEY="your-api-key"

Get your API key from the Synthflow Dashboard or use the setup-api-key skill.

Quick Start

1. Get an API key: Use the setup-api-key skill or visit https://app.synthflow.ai
2. Create an assistant: Use the create-assistant skill to build a voice AI agent
3. Add actions: Use manage-actions to add transfers, SMS, webhooks, and more
4. Make a call: Use create-call to test your assistant

API Reference

• Base URL: https://api.synthflow.ai/v2
• Authentication: Bearer token via Authorization: Bearer $SYNTHFLOW_API_KEY
• Full API Docs: https://docs.synthflow.ai
• MCP Docs Server: https://docs.synthflow.ai/_mcp/server (auto-configured via .mcp.json)

Telemetry

These skills collect anonymous usage data via PostHog to help us improve. We track:

• README views — a tracking pixel fires when the README is viewed on GitHub
• Skill usage — a lightweight event is sent when an agent completes a skill workflow

What's collected: event name, skill name, success/failure, and a hashed user identifier (SHA-256 of $USER). No personal information is sent.

Opt out: Set either environment variable to disable all telemetry:

export DO_NOT_TRACK=1
# or
export DISABLE_TELEMETRY=1