The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the opentaint test approximation run command to validate generated Java approximation sources. This is a core part of the developer workflow and operates on local project directories using defined input variables.
[PROMPT_INJECTION]: The skill processes external configuration (dropped-external-methods.yaml) and library source code to inform its code generation tasks. This represents an indirect prompt injection surface where instructions hidden in the library source or configuration could theoretically influence the agent's output. However, the risk is minimal as the agent is constrained to generating structured Java models for a specific security analysis tool.

create-dataflow-approximation