The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the dxs command-line tool to perform authorized actions such as checking authentication status, listing organization connections, and querying OData schema information. The commands are scoped to the dxs platform and are used according to standard development practices for schema discovery.
[INDIRECT_PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection as it ingests untrusted metadata from external OData services. 1. Ingestion points: Data enters the agent context through dxs schema search, describe-entity, and other schema discovery command outputs. 2. Boundary markers: None identified, but instructions focus on generating a static field-mapping.md report. 3. Capability inventory: The skill uses dxs CLI commands for metadata retrieval and does not expose high-privilege system operations to the ingested data. 4. Sanitization: The skill produces markdown artifacts for human review, reducing the risk of automated execution of malicious instructions embedded in the metadata.

schema-explorer