The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted data from external sources, creating a potential surface for indirect prompt injection attacks.
Ingestion points: The skill processes user-supplied text messages and external images provided via the image_urls parameter in API requests (found in SKILL.md).
Boundary markers: There are no explicit instructions or delimiters defined to help the agent distinguish between its core instructions and potentially malicious commands embedded within the input data or images.
Capability inventory: The skill facilitates network operations by communicating with the EachLabs API endpoint (eachsense-agent.core.eachlabs.run) to perform render generations.
Sanitization: The documentation does not specify any validation, filtering, or escaping mechanisms for the text descriptions or the content of the external images being processed.
[EXTERNAL_DOWNLOADS]: The skill documentation and examples reference external network resources for API functionality and asset storage.
The skill communicates with eachsense-agent.core.eachlabs.run for task execution and references storage.eachlabs.ai for output retrieval. These domains are identified as official infrastructure belonging to the vendor 'eachlabs'.
One example references example.com as a placeholder for architectural sketches.

architecture-rendering