The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to use curl commands to interact with various Opper API endpoints (e.g., /v3/images, /v3/audio/speech, /v3/ocr). These commands are used to transmit data to and receive media from the vendor's infrastructure.
[EXTERNAL_DOWNLOADS]: The skill directs the agent to download the live OpenAPI specification from https://api.opper.ai/v3/openapi.yaml to ensure accurate usage of the API. This network activity targets the vendor's official technical documentation.
[PROMPT_INJECTION]: The skill facilitates the processing of external content such as PDFs (via OCR), audio recordings (via transcription), and images (via vision models). This creates a vulnerability surface for indirect prompt injection attacks.
Ingestion points: External data is ingested through endpoints defined in SKILL.md such as POST /v3/ocr and POST /v3/audio/transcriptions.
Boundary markers: The instructions do not define delimiters or specific warnings to ignore instructions embedded within the processed media.
Capability inventory: The skill provides capabilities for network communication (via curl) and remote file management on the Opper platform as described in SKILL.md.
Sanitization: There are no documented procedures for sanitizing or filtering the content extracted from external media.

opper-multimodal