The Agent Skills Directory

[COMMAND_EXECUTION]: The skill provides example curl commands for users to interact with the API endpoints directly from their terminal.
[DATA_EXFILTRATION]: The skill transmits data and potentially sensitive metadata to api.iblai.app. This is legitimate functional behavior as it targets the official domain of the skill author (iblai) and uses standard authentication via environment variables.
[INDIRECT_PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection as it processes external metadata and file content.
Ingestion points: Metadata fields in API responses such as title, description, and file_url, as well as user-provided binary file data.
Boundary markers: The instructions do not define specific delimiters or instructions to ignore embedded content in the ingested data.
Capability inventory: The skill performs network operations via curl and handles multipart file uploads to the Data Manager API.
Sanitization: There is no explicit mention of sanitization, filtering, or validation for the metadata strings processed at runtime.

iblai-api-catalog-media