Use when user needs to convert audio or voice recordings to text. Provides necessary command usage instructions for converting audio to text. Supports multip...
Security Analysis
medium confidenceThe skill's instructions coherently describe a speech-to-text CLI, but the package metadata omits the required binary and has no source/homepage — this mismatch and lack of provenance warrant caution.
The SKILL.md explicitly instructs running the external command 'miaoda-studio-cli speech-to-text', which is necessary for the described purpose. However, the skill metadata lists no required binaries and provides no install spec or homepage/source. That omission is an inconsistency: a speech-to-text skill legitimately needs the CLI it documents, so the metadata should declare that dependency or provide an install/source.
The instructions themselves stay on-topic (describe command flags, examples, language options, and common errors) and only refer to operating on a user-supplied audio file. They do not request extra environment variables or access to unrelated files. The SKILL.md does reference another skill ('miaoda-web-search') for a follow-on action, which is transparent. The main issue is that the instructions assume the presence of an external binary that the skill does not declare or provide.
There is no install spec and no code files — the skill is instruction-only. That reduces the surface written to disk by the skill bundle itself. The remaining risk comes from whatever 'miaoda-studio-cli' the user or agent runs, which is outside the skill package.
The skill requests no environment variables or credentials in metadata and the instructions do not reference secrets or unrelated config. This is proportionate to a CLI-guidance skill.
The skill does not request always:true and is user-invocable only. It does not declare any capability to modify other skills or agent-wide settings. Autonomous invocation remains possible (platform default) but is not combined with other concerning privileges here.
Guidance
This skill is an instruction-only guide that tells the agent to run 'miaoda-studio-cli speech-to-text', but the skill metadata gives no install instructions or homepage. Before using: verify what 'miaoda-studio-cli' is and where it comes from (official docs, package repository, signed release). Do not run unknown binaries on sensitive data or systems. If you plan to use it, prefer installing the CLI from an official source, check signatures/reviews, and test on non-sensitive audio first. If you cannot verify the CLI's origin, avoid granting the agent access to files you care about or decline to install the skill.
Latest Release
v1.0.0
Initial release of miaoda-speech-to-text. - Converts audio or voice recordings to text using CLI commands. - Supports multiple languages including Chinese, English, French, Spanish, Japanese, and more. - Offers detailed usage instructions, CLI parameter guide, and usage examples. - Provides tips for common scenarios and common mistakes to avoid. - Output formats supported: plain text and JSON.
Popular Skills
Published by @nice1234-h on ClawHub
