The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes potentially untrusted data from external sources, creating a surface for indirect prompt injection.
Ingestion points: The skill uses model_read to ingest controller model data and read_web_page to ingest external domain research from the internet.
Boundary markers: The instructions do not specify any boundary markers or delimiters to isolate ingested data from the agent's primary instructions.
Capability inventory: The skill utilizes web_search and read_web_page for network access, and model_test for executing simulations.
Sanitization: No sanitization or validation of the external content is specified in the workflow.

specifying-plant-models