← Back to Skills Marketplace
2485
Downloads
3
Stars
3
Active Installs
5
Versions
Install in OpenClaw
/install openclaw-voiceai-voice-agent
Description
Create, manage, and deploy Voice.ai conversational AI agents. Use when the user wants to work with voice agents, list their agents, create new ones, or manage agent configurations.
Usage Guidance
This skill appears to do what it says (manage Voice.ai agents) but its metadata is incomplete and could lead to surprising behavior. Before installing: (1) review the included JavaScript files (voice-ai-agents-sdk.js and scripts/agent.js) to see what network endpoints they call, whether they log or transmit data, and whether they use any other env vars or child processes; (2) ensure you have a dedicated API key with minimal scopes for this skill rather than a broadly privileged key; (3) don't commit a .env with secrets into source control — prefer OpenClaw config or a secrets manager; (4) if you don't want the agent to be triggered autonomously, set disable-model-invocation or otherwise restrict automatic triggers; (5) ask the publisher for a homepage/source repo and an explanation of required Node version and dependency list. If you can't review the code or verify the publisher, treat the skill as untrusted.
Capability Analysis
Type: OpenClaw Skill
Name: openclaw-voiceai-voice-agent
Version: 0.1.4
The skill bundle provides a legitimate client for the Voice.ai API, enabling management of conversational AI agents. All network calls are directed to the official `dev.voice.ai` endpoint, and API key handling is standard via environment variables. The `SKILL.md` includes an `echo $VOICE_AI_API_KEY` command, but this is presented as a verification step for the agent, not an instruction for exfiltration. There is no evidence of malicious execution, persistence, obfuscation, or prompt injection attempts to subvert the OpenClaw agent's core function.
Capability Assessment
Purpose & Capability
Name and description align with the included SDK and CLI scripts: creating, listing, deploying voice agents is coherent. However, the registry metadata declares no required environment variables or primary credential even though SKILL.md instructs the user to set VOICE_AI_API_KEY. Also the skill has no documented homepage/source to verify origin.
Instruction Scope
SKILL.md directly instructs users to set VOICE_AI_API_KEY and to run node scripts (scripts/agent.js). The runtime instructions implicitly require a Node.js runtime and network access to Voice.ai endpoints. The docs also claim automatic OpenClaw activation on certain chat triggers, but the registry flags for invocation/persistence were not explicitly set — an inconsistency. The instructions tell users to create/append a .env file (which could leak if stored in repo) and to export the API key into the environment.
Install Mechanism
No install spec is provided (instruction-only), yet the package contains two JS files (SDK and CLI). The skill requires Node to execute scripts, but required binaries are declared as none in metadata. That mismatch is important: executing the shipped JS will write network calls and possibly perform I/O, but there is no declaration of runtime dependencies or vetted install source.
Credentials
SKILL.md requires a VOICE_AI_API_KEY for operation, but the registry metadata lists no required env vars or primary credential. The skill also suggests multiple ways to store the key (env, .env, OpenClaw config) but doesn't state what scopes/permissions are necessary. Because model invocation is allowed by default, an API key supplied to the environment could be used autonomously by the agent unless the operator restricts invocation.
Persistence & Privilege
The skill does not set always:true (good) but also does not set disable-model-invocation, so the model can invoke the skill autonomously. Combined with the (undeclared) credential the skill needs, that creates a risk that the model could call external APIs using your key without explicit user action. The skill's metadata lacks explicit statements about when and how credentials are read.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install openclaw-voiceai-voice-agent - After installation, invoke the skill by name or use
/openclaw-voiceai-voice-agent - Provide required inputs per the skill's parameter spec and get structured output
Version History
v0.1.4
- Major documentation update: SKILL.md now covers API key setup, authentication methods, error handling, and OpenClaw integration in detail.
- Expanded documentation for configuration via environment variable, .env file, and OpenClaw config.
- Added comprehensive CLI, JSON configuration, and user-trigger phrase reference tables.
- Improved troubleshooting section with common errors and solutions.
- Included more usage examples, file overviews, and links for quick onboarding.
v0.1.3
- Minor update to documentation for clarity and improved wording in SKILL.md.
- Enhanced the introduction to emphasize building the best conversational AI voice agents.
- No changes to features, configuration, or usage instructions.
v0.1.2
- Skill and file names updated from "voice.ai-agents" to "voice.ai-voice-agents" for clarity and consistency.
- Documentation (SKILL.md) reflects the new name throughout.
- No changes to functionality or usage instructions.
v0.1.1
- Fixed inconsistent skill name in metadata front matter (now "voice.ai-agents")
- Added a "title" field to metadata for improved clarity and presentation
v0.1.0
Initial release introducing Voice.ai conversational agent management.
- Create, update, deploy, pause, and delete voice agents via CLI.
- RAG-powered knowledge base support for custom agent information.
- MCP integration for connecting agents to external tools.
- Full configuration options for LLM, TTS voice, and call settings.
- Built-in analytics and phone number management features.
Metadata
Frequently Asked Questions
What is Voice.Ai Voice Agents?
Create, manage, and deploy Voice.ai conversational AI agents. Use when the user wants to work with voice agents, list their agents, create new ones, or manage agent configurations. It is an AI Agent Skill for Claude Code / OpenClaw, with 2485 downloads so far.
How do I install Voice.Ai Voice Agents?
Run "/install openclaw-voiceai-voice-agent" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Voice.Ai Voice Agents free?
Yes, Voice.Ai Voice Agents is completely free (open-source). You can download, install and use it at no cost.
Which platforms does Voice.Ai Voice Agents support?
Voice.Ai Voice Agents is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Voice.Ai Voice Agents?
It is built and maintained by Nick Gill (@gizmogremlin); the current version is v0.1.4.
More Skills