npx claudepluginhub voicetestdev/voicetest --plugin voicetestThis skill uses the workspace's default tool permissions.
voicetest is a CLI tool and library for testing voice agent workflows across platforms.
Builds ElevenLabs conversational AI voice agents: configure via CLI/dashboard, add tools/knowledge, integrate React/React Native/Swift/JS SDKs, test/deploy. For voice AI, phone systems, or ElevenLabs errors.
Guides building voice AI agents with LiveKit Cloud and Agents SDK, including project setup, LiveKit Inference integration, workflows, handoffs, and mandatory testing.
Build voice AI agents with LiveKit Agents SDK using Cloud Inference or self-hosted setups and lk CLI. Includes checklists for credentials, docs, and testing.
Share bugs, ideas, or general feedback.
voicetest is a CLI tool and library for testing voice agent workflows across platforms. It imports agent definitions, runs simulated conversations, evaluates results against metrics, and exports to different formats.
import → test → evaluate → export
voicetest run -a agent.json -t tests.json --all # Run all tests
voicetest run -a agent.json -t tests.json --test "Name" # Run specific test
voicetest --json run -a agent.json -t tests.json --all # JSON output for parsing
voicetest agent list # List agents in database
voicetest agent get <id> # Get agent details
voicetest agent create -a agent.json # Create from file
voicetest agent update <id> --name "New Name" # Update properties
voicetest agent delete <id> # Delete agent
voicetest agent graph <id> # Show graph structure
voicetest test list <agent-id> # List test cases
voicetest test create <agent-id> -f test.json # Create from file
voicetest test link <agent-id> tests.json # Link external file
voicetest test export <agent-id> # Export all tests
voicetest runs list <agent-id> # List past runs
voicetest runs get <run-id> # Get run details
voicetest snippet analyze --agent agent.json # Find repeated text
voicetest snippet list --agent agent.json # List defined snippets
voicetest snippet set --agent agent.json greeting "Hello!" # Set snippet
voicetest export -a agent.json -f mermaid # Mermaid diagram
voicetest export -a agent.json -f livekit # LiveKit Python agent
voicetest export -a agent.json -f vapi # VAPI format
voicetest importers # List importers
voicetest exporters # List export formats
voicetest platforms # List platforms
voicetest platform configure retell --api-key <key> # Set credentials
voicetest platform list-agents retell # List remote agents
voicetest platform import retell <agent-id> -o agent.json # Import agent
voicetest platform push retell -a agent.json # Push to platform
voicetest evaluate -t transcript.json -m "Agent was polite" # Evaluate transcript
voicetest diagnose -a agent.json -t tests.json --auto-fix # Auto-fix failures
voicetest chat -a agent.json # Interactive chat
voicetest settings # Show settings
voicetest settings --set models.agent=openai/gpt-4o # Set a value
voicetest settings --defaults # Show defaults
All commands support --json for machine-parseable output. Progress messages go to
stderr, structured data goes to stdout.
voicetest --json agent list | jq '.[].name'
voicetest --json run -a agent.json -t tests.json --all | jq '.results[] | {name: .test_name, status}'
voicetest uses three LLM roles:
Configure via .voicetest.toml or CLI:
voicetest settings --set models.agent=openai/gpt-4o
voicetest settings --set models.judge=anthropic/claude-sonnet-4-20250514
See references/test-format.md for the full JSON schema.
See references/platforms.md for platform-specific notes.