Percept Voice Cmd
Skill by jarvis563
Why use this skill?
Enable voice control for your OpenClaw agent with Percept Voice Cmd. Manage emails, notes, and reminders hands-free with intelligent wake-word detection.
Install via CLI (Recommended)
clawhub install openclaw/skills/skills/jarvis563/percept-voice-cmdWhat This Skill Does
The Percept Voice Cmd skill, developed by jarvis563, serves as a high-performance bridge between ambient audio environments and your OpenClaw agent. It functions by continuously monitoring incoming audio streams for specific trigger phrases, commonly known as wake words. Once a wake word is detected, the skill initiates a robust capture window, extending the recording period to ensure that complex multi-word commands are captured in their entirety. It utilizes a sophisticated two-tier parsing system, beginning with lightning-fast regex matching for common routine tasks and falling back to LLM-driven inference for more nuanced or conversational requests. By integrating tightly with OpenClaw’s command-line interface, it translates spoken intent into concrete actions, such as sending communications, setting schedule alerts, or performing web-based queries.
Installation
To integrate this skill into your environment, ensure you have the percept-listen dependency running, as this is required for the audio buffering and wake word detection pipeline. Once the prerequisite is met, execute the following command in your terminal: clawhub install openclaw/skills/skills/jarvis563/percept-voice-cmd. After installation, navigate to the local Percept dashboard at port 8960 to configure your preferred wake words and authorize specific speaker profiles, which prevents unauthorized audio input from triggering your agent’s actions.
Use Cases
This skill is designed for scenarios requiring hands-free agent interaction. Use it when your hands are occupied, such as while cooking or working on a hardware project, to jot down notes or check your calendar. It is also ideal for office environments where you need to send quick emails or set reminders without interrupting your workflow to use a keyboard. The speaker authorization feature ensures that this tool remains secure in shared spaces by restricting control to verified users only.
Example Prompts
- "Hey Jarvis, send an email to Sarah regarding the project kickoff meeting on Monday."
- "Hey Jarvis, remind me in fifteen minutes to pull the laundry out of the dryer."
- "Hey Jarvis, what does my schedule look like for the rest of the day?"
Tips & Limitations
To optimize performance, place your microphone in a central location and refine your percept/data/contacts.json file regularly to ensure accurate name resolution. Be aware that the skill relies on your local OpenClaw agent instance being active and reachable via CLI. For privacy, always review the percept/data/speakers.json file periodically to confirm that only authorized individuals have permission to execute commands through your system.
Metadata
Not sure this is the right skill?
Describe what you want to build — we'll match you to the best skill from 16,000+ options.
Find the right skillPaste this into your clawhub.json to enable this plugin.
{
"plugins": {
"official-jarvis563-percept-voice-cmd": {
"enabled": true,
"auto_update": true
}
}
}Tags(AI)
Flags: file-read, file-write, code-execution
Related Skills
Percept Ambient
Skill by jarvis563
Percept Summarize
Skill by jarvis563
browser-audio-capture
Capture audio from any browser tab — meetings, YouTube, podcasts, courses, webinars — and stream to any AI agent. Zero API keys, works with any framework.
Percept Speaker Id
Skill by jarvis563
stable-browser
Set up reliable browser automation using Chrome DevTools Protocol (CDP) instead of the flaky browser extension relay. Use when browser relay keeps disconnecting, throwing WebSocket 403 errors, or when you need stable headless/headed browser control for web scraping, form filling, social media posting, or any browser automation task. Replaces profile="chrome" with a rock-solid CDP connection.