The Agent Skills Directory

[PROMPT_INJECTION]: The skill contains explicit instructions to bypass content safety filters by setting the enable_safety_checker parameter to false in API requests. This guides the agent to circumvent standard AI safety guardrails for the purpose of generating adult content.
[DATA_EXFILTRATION]: The skill performs network operations by sending user-provided prompts and image URLs to an external API endpoint (https://sense.eachlabs.run/chat). While this is a vendor-owned resource, it involves transmitting data to a remote server.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted input from users that could contain malicious instructions designed to manipulate the agent's behavior or the video generation output.
Ingestion points: User prompts within the messages object and external image links provided in the image_urls parameter.
Boundary markers: The skill lacks delimiters or explicit instructions to the agent to treat the user-supplied content as data rather than instructions.
Capability inventory: The skill has the capability to perform network requests to external domains and process remote media files.
Sanitization: No input validation, escaping, or filtering mechanisms are described to mitigate risks from malicious user content.
[EXTERNAL_DOWNLOADS]: The documentation references external dependencies for processing API responses, specifically the eventsource Node.js package and the requests Python library. These are well-known packages used for handling streaming data and HTTP requests.

NSFW Video Generation