The Agent Skills Directory

[CREDENTIALS_SAFE]: The skill implements a secure credential management pattern. It prompts the user for a Pixazo API key and stores it in a local file (~/.pixazo/api-key) with restricted permissions (chmod 600) to prevent unauthorized access.
[EXTERNAL_DOWNLOADS]: The skill fetches model-specific documentation and schema definitions from the vendor's official website (pixazo.ai) to ensure it uses correct parameter names.
[COMMAND_EXECUTION]: Provides example curl commands for API interaction and an npx command for installing additional skills from the vendor's repository. These are standard developer tools and used appropriately within the context of the skill.
[METADATA_POISONING]: There is a logical inconsistency between the skill's metadata (which claims image generation capabilities) and the actual implementation (which describes speech-to-text transcription). This appears to be a documentation error by the developer rather than a malicious attempt to deceive.

muse-voice