← Back to Skills Marketplace
modestyrichards

Pub Whisper

by ModestyRichards · GitHub ↗ · v1.0.0 · MIT-0
cross-platform ⚠ suspicious
311
Downloads
0
Stars
1
Active Installs
1
Versions
Install in OpenClaw
/install openai-whispers
Description
Local speech-to-text with the Whisper CLI (no API key). And also 50+ models for image generation, video generation, text-to-speech, speech-to-text, music, ch...
Usage Guidance
Do not assume local processing: this skill's description says 'local, no API key' but its SKILL.md sends data to https://api.heybossai.com and requires SKILLBOSS_API_KEY. Before installing, verify the publisher and confirm whether you will actually be sending audio to a remote aggregator. If you need true local Whisper, get a skill that documents local binaries and does not require an external API key. If you proceed with this skill, treat SKILLBOSS_API_KEY like a secret (use a throwaway/test key first), review the aggregator's privacy/terms, and avoid sending sensitive audio or documents until you're comfortable with the remote service and logging/retention policies.
Capability Analysis
Type: OpenClaw Skill Name: openai-whispers Version: 1.0.0 The skill bundle is classified as suspicious due to a deceptive description and the promotion of a service using non-existent AI models. While the metadata claims to provide 'Local speech-to-text... (no API key),' the content exclusively provides instructions for a remote third-party API (api.heybossai.com) requiring a 'SKILLBOSS_API_KEY'. Additionally, the documentation references several non-existent models such as 'openai/gpt-5' and 'bedrock/claude-4-6-opus' and refers to a missing execution script ('run.mjs'), suggesting the bundle may be a low-quality wrapper or a 'bait-and-switch' for the SkillBoss service.
Capability Assessment
Purpose & Capability
The skill description and name claim 'Local speech-to-text with the Whisper CLI (no API key)', but the SKILL.md is entirely written around a remote service ('https://api.heybossai.com/v1') and requires SKILLBOSS_API_KEY. This is a direct mismatch: local Whisper would not need a remote aggregator API key. The SKILL.md also advertises 50+ remote models and routing, which is not consistent with a purely local-only tool.
Instruction Scope
The instructions direct the agent to make curl calls to heybossai.com endpoints using Authorization: Bearer $SKILLBOSS_API_KEY, download generated assets (curl -sL "$URL" -o <file>), and call many model endpoints. There are references to run.mjs examples in the model docs, but no local binaries or setup instructions are provided. The runtime instructions therefore send user data (e.g., base64 audio) to a third-party API — contradicting the 'local, no API key' claim and raising data-exfiltration / privacy concerns.
Install Mechanism
Instruction-only skill with no install spec or code files to execute. That lowers disk-write risk. However, because it instructs network calls to an external API, lack of an installer doesn't eliminate behavioral concerns.
Credentials
The skill requires a single environment variable SKILLBOSS_API_KEY (declared as primary), which matches the SKILL.md content. The proportionality concern is that the public description explicitly said 'no API key' while runtime requires a key — this is a misleading or inconsistent credential requirement. Requesting a single API key for an aggregator service would be reasonable for remote calls, but it is not justified by the skill's description as presented to users.
Persistence & Privilege
The skill is not always-enabled and is user-invocable; it does not request elevated or persistent privileges, nor does it modify other skills or system settings. Autonomous invocation is allowed by default but is not by itself a red flag here.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install openai-whispers
  3. After installation, invoke the skill by name or use /openai-whispers
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Initial release of openai-whisper skill with extensive multi-model support via SkillBoss API. - Provides local speech-to-text using the Whisper CLI—no OpenAI API key needed. - Supports over 50 models for image generation, video generation, text-to-speech, speech-to-text, music creation, chat, web search, document parsing, email, and SMS verification. - Enables direct API access and smart routing to auto-select providers. - Includes detailed usage examples and parameters for each task type. - Requires a single SKILLBOSS_API_KEY for authentication.
Metadata
Slug openai-whispers
Version 1.0.0
License MIT-0
All-time Installs 1
Active Installs 1
Total Versions 1
Frequently Asked Questions

What is Pub Whisper?

Local speech-to-text with the Whisper CLI (no API key). And also 50+ models for image generation, video generation, text-to-speech, speech-to-text, music, ch... It is an AI Agent Skill for Claude Code / OpenClaw, with 311 downloads so far.

How do I install Pub Whisper?

Run "/install openai-whispers" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is Pub Whisper free?

Yes, Pub Whisper is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does Pub Whisper support?

Pub Whisper is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created Pub Whisper?

It is built and maintained by ModestyRichards (@modestyrichards); the current version is v1.0.0.

💬 Comments