HiveMind CLI

The simplest HiveMind satellite — text only, no audio hardware required.

What runs locally: A CLI interface. No audio processing whatsoever.

What the hub provides: All processing (STT, TTS, skills, intents). Responses arrive as text.

When to use it

• Testing and debugging a hub or a skill
• Scripting and automation from a server
• IoT devices with network connectivity but no audio hardware
• Any scenario where you want to interact via keyboard or piped input

Install

pip install hivemind-cli

Usage

Interactive mode — type utterances and see responses:

hivemind-client --host 192.168.1.10 --key <key> --password <password>

Or with an identity file set via hivemind-client set-identity:

hivemind-client

Send a single utterance and exit:

hivemind-client send-utterance "what time is it"

Send a raw OVOS message:

hivemind-client send-mycroft --msg "speak" --payload '{"utterance": "hello world"}'

Other CLI commands

# Escalate a message up the hive chain
hivemind-client escalate --msg "recognizer_loop:utterance" --payload '{"utterances": ["hello"]}'

# Propagate a message to all nodes
hivemind-client propagate --msg "recognizer_loop:utterance" --payload '{"utterances": ["hello"]}'

# Test connection from the stored identity file
hivemind-client test-identity

# Set the identity file
hivemind-client set-identity \
  --key <key> --password <password> --host <host> --port 5678 --siteid myroom

# Reset PGP keys
hivemind-client reset-pgp

Run hivemind-client --help for a full list.