The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes curl commands to interact with the Lytics API at api.lytics.io. These operations are used for discovering providers, checking authentication, and managing integration jobs.
[DATA_EXFILTRATION]: Authentication is handled via the LYTICS_API_TOKEN environment variable. All network requests are directed to the vendor's official API domain, and no sensitive local files or credentials are exfiltrated to external parties.
[PROMPT_INJECTION]: The skill demonstrates an indirect prompt injection surface by processing data from external connections (via schema discovery and data sampling) to suggest field mappings. This is managed by a structured confirmation step. 1. Ingestion points: SKILL.md Step 5 calls to /v2/connection/${CONNECTION_ID}/schema and /v2/connection/${CONNECTION_ID}/scan. 2. Boundary markers: None explicitly defined in the agent instructions, though the workflow includes a confirmation-gate pattern. 3. Capability inventory: The skill uses curl to perform network operations and resource creation via the Lytics API. 4. Sanitization: The skill does not explicitly mention sanitization of the external data used for generating mapping recommendations.

integration-advisor