The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes external media (audio, video, image) and text content, which serves as a potential vector for indirect prompt injection where malicious instructions could be embedded in the data being analyzed.
Ingestion points: Media files provided via URLs in the url parameter for the /detect and /intelligence endpoints, and raw text input for the /text_detect endpoint.
Boundary markers: There are no explicit instructions or delimiters defined to guide the agent in distinguishing between data to be analyzed and potential embedded instructions within that data.
Capability inventory: The skill interacts with the Resemble AI API and utilizes various MCP tools (resemble_docs_lookup, resemble_search, resemble_api_endpoint, resemble_api_search, resemble_get_page, resemble_list_topics) to perform its tasks.
Sanitization: The skill does not specify any sanitization, validation, or filtering of the content retrieved from external URLs or provided in text inputs.

resemble-detect