The Agent Skills Directory

[COMMAND_EXECUTION]: The skill provides detailed instructions and examples for executing the opentaint CLI tool, including the use of flags that enable the automatic compilation and loading of user-provided Java source files for security analysis.
[SAFE]: The skill implements a workflow where data from a previous analysis report (external-methods-without-rules.yaml) is ingested to drive the creation of new Java source files. This pattern is a standard requirement for security analysis tasks, although it represents a potential indirect prompt injection surface if the input data were to be maliciously manipulated.
Ingestion points: Reads class and method names from external-methods-without-rules.yaml.
Boundary markers: None specified in the instructions.
Capability inventory: Writing Java source files to the file system and executing the opentaint CLI.
Sanitization: No explicit sanitization or validation of the ingested method data is described.

create-approximation