The Agent Skills Directory

[PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection by processing external data from a project's model and compiled classes. Ingestion points: The skill ingests data from the project's project.yaml file and bytecode from .class files within the provided <model-dir> via javap output. Boundary markers: No explicit delimiters or instructions are provided to the agent to treat the analyzed code or configuration as untrusted data. Capability inventory: The skill has the capability to execute shell scripts and PowerShell scripts, invoke JDK tools like javap and jar, and perform file-system operations (read/write). Sanitization: The skill does not implement specific sanitization or validation routines for the data extracted from the project files before processing it.

discover-attack-surface