The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests data from external sources that could contain malicious instructions.
Ingestion points: The skill retrieves message bodies, homework descriptions, and schedule details from the Magister API in magister.mjs.
Boundary markers: There are no explicit delimiters or instructions provided in the output to ensure the agent ignores instructions embedded within the school data.
Capability inventory: The skill includes the report-sick command which performs a state-changing POST request to the API.
Sanitization: The htmlToText function in magister.mjs provides basic sanitization by removing HTML tags from the ingested content.
[COMMAND_EXECUTION]: The skill executes a local Node.js script (magister.mjs) to interact with the service.
[DATA_EXFILTRATION]: The skill performs network operations to external domains to retrieve student data.
Evidence: Network requests are made to accounts.magister.net and *.magister.net hosts as required for its primary functionality.

magister