The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes curl to perform API calls to the MiMo search service. This is the intended behavior for the functionality described.
[EXTERNAL_DOWNLOADS]: Retrieves search results and citations from api.xiaomimimo.com. The domain belongs to a well-known technology company and is a legitimate source for the service.
[DATA_EXFILTRATION]: Authentication is handled using the MIMO_API_KEY environment variable. This is a secure and standard practice for providing API credentials to agents without hardcoding them in the skill source.
[PROMPT_INJECTION]: The skill has a surface for indirect prompt injection as it ingests untrusted data from the internet.
Ingestion points: Data enters the agent context through the MiMo API search results (SKILL.md).
Boundary markers: The API uses a structured JSON response with annotations for citations, providing a clear boundary between search data and model logic.
Capability inventory: The skill is limited to performing network requests via curl and does not have local file system write or code execution capabilities.
Sanitization: The skill relies on the structured API response format for data integrity.

mimo-web-search