The Agent Skills Directory

[Indirect Prompt Injection Surface]: The skill is designed to retrieve and process records from Microsoft Dataverse, which acts as an external data source. This is a primary function of the skill, and users should be aware that the data retrieved enters the agent's context.
Ingestion points: External data enters the agent context through client.records.get, client.dataframe.get, and raw Web API calls (via urllib.request) defined in the SKILL.md file.
Boundary markers: The instructions do not define specific delimiters or instructions to treat retrieved record data as untrusted content.
Capability inventory: The skill provides capabilities for Python execution, file system operations (writing CSV files), and network communication with Dataverse endpoints.
Sanitization: There are no documented sanitization or validation steps for the content retrieved from Dataverse before it is processed or presented to the user.
[Credential Management Pattern]: The skill utilizes environment variables and local module imports (scripts/auth.py) for authentication and configuration. This follows common patterns for managing service URLs and access credentials securely outside of the main instruction logic.
[Network Connectivity]: The skill performs network operations to Dataverse API endpoints to retrieve data and execute server-side aggregations. These operations are directed toward Microsoft's official infrastructure as part of the skill's intended data access workflow.

dv-query