The Agent Skills Directory

[SAFE]: The skill contains no executable code or scripts. It is a documentation-only skill providing guidance to an AI agent on how to interact with Azure Speech APIs. All technical information is descriptive and instructional.
[EXTERNAL_DOWNLOADS]: The skill provides numerous links to external documentation. All URLs target the official microsoft.com and azure.microsoft.com domains, which are recognized as trusted, well-known services. These references are used legitimately to cite service limits, pricing, and technical overviews.
[PROMPT_INJECTION]: There are no attempts to override the agent's system instructions or bypass safety filters. The instructions explicitly direct the agent to uphold ethical standards, such as refusing to clone voices without explicit talent consent and disclosing the synthetic nature of generated content.
[INDIRECT_PROMPT_INJECTION]: The skill describes processes that involve ingest of external data (audio and video files). It includes mitigation guidance to handle these inputs safely, such as utilizing 'Responsible AI' transparency notes, requesting user confirmation before taking significant actions, and using phrase lists to improve accuracy without compromising security boundaries.

azure-speech