← Back to Skills Marketplace
Discord Voice
by
avatarneil
· GitHub ↗
· v0.1.6
6496
Downloads
8
Stars
33
Active Installs
6
Versions
Install in OpenClaw
/install discord-voice
Description
Real-time voice conversations in Discord voice channels with Claude AI
Usage Guidance
Review before installing on shared servers. Restrict allowedUsers, avoid auto-joining public channels, tell participants that audio/transcripts may be processed by external providers, prefer local STT/TTS where possible, and ensure the underlying OpenClaw agent cannot perform sensitive actions from voice prompts without approval.
Capability Analysis
Type: OpenClaw Skill
Name: discord-voice
Version: 0.1.6
The OpenClaw Discord Voice plugin is designed with strong security considerations. It implements robust sanitization for all user-controlled inputs before they are used in agent prompts or file paths, specifically preventing prompt injection via user IDs and path traversal for custom sound files. The dynamic loading of the core OpenClaw extension API is performed with integrity checks, verifying the package name and resolving paths securely. While the plugin connects to various external STT/TTS APIs and a configurable local Wyoming Whisper server, these network interactions are for its stated purpose and API keys are handled via configuration or environment variables, not directly from user input. No evidence of intentional harmful behavior, such as data exfiltration to unauthorized endpoints or backdoor installation, was found.
Capability Assessment
Purpose & Capability
The stated purpose matches the code: it joins Discord voice channels, records speech, transcribes it, runs an OpenClaw agent, and speaks responses. The concern is that this grants live Discord speakers a voice path into the normal agent capabilities, not just voice-specific behavior.
Instruction Scope
The embedded agent prompt explicitly says the agent has access to all normal tools and skills, and the code does not show a voice-lane tool allowlist or confirmation gate for sensitive actions triggered by transcripts.
Install Mechanism
Installation is a standard Node/OpenClaw plugin flow with declared dependencies and no postinstall script observed. The dynamic OpenClaw API load includes package-name checks and is documented.
Credentials
Discord, STT, LLM, and TTS network access are purpose-aligned, but default cloud providers, transcript logging, and an empty allowedUsers default create privacy and consent exposure for people in joined voice channels.
Persistence & Privilege
The plugin registers a background service, keeps Discord voice sessions alive with heartbeat/reconnect logic, and may persist OpenClaw session entries. Auto-join is configurable rather than hidden, but operators should treat it as always-on voice infrastructure when enabled.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install discord-voice - After installation, invoke the skill by name or use
/discord-voice - Provide required inputs per the skill's parameter spec and get structured output
Version History
v0.1.6
discord-voice 0.1.6
- No code or documentation changes detected in this version.
- Functionality and configuration remain identical to the previous release.
v0.1.5
Resolve transient package vulns (tar).
v0.1.4
- Added continuous integration workflow: `.github/workflows/ci.yml`.
- Introduced code formatting and linting configuration files: `.oxfmtrc.json` and `.oxlintrc.json`.
- Added a security policy: `SECURITY.md`.
- Documentation update: Corrected Discord command and CLI usage instructions to use `discord_voice` instead of `voice` for consistency.
v0.1.2
- Changed config metadata: now uses `requiredConfig` with `discord.token` instead of `requiredEnv` for `DISCORD_TOKEN`.
- No code or logic changes; only SKILL.md metadata updated.
v0.1.1
- Added support for "local-whisper" as a Speech-to-Text provider, allowing offline transcription.
- Updated documentation to reflect new configuration defaults and provider options (default `sttProvider` is now "local-whisper").
- Added new files: openclaw.plugin.json, scripts/smoke-test.mjs, src/constants.ts.
v0.1.0
Initial release
Metadata
Frequently Asked Questions
What is Discord Voice?
Real-time voice conversations in Discord voice channels with Claude AI. It is an AI Agent Skill for Claude Code / OpenClaw, with 6496 downloads so far.
How do I install Discord Voice?
Run "/install discord-voice" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Discord Voice free?
Yes, Discord Voice is completely free (open-source). You can download, install and use it at no cost.
Which platforms does Discord Voice support?
Discord Voice is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Discord Voice?
It is built and maintained by avatarneil (@avatarneil); the current version is v0.1.6.
More Skills