The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes standard Maven commands (mvn compile, mvn verify) to ensure project stability before and after implementing testing changes. It also executes Docker commands to build and run the fuzzer image.
[DYNAMIC_EXECUTION]: The instructions direct the agent to generate a run-cats-fuzz.sh script from an internal asset and execute it. This script handles the orchestration of the Docker-based fuzzing process. The script includes safety features like path validation to ensure files remain within the repository root, preventing unauthorized file access.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to process external OpenAPI specification files provided by the user. While this represents a data ingestion surface, the skill implements a 'blocking safety check' requiring user input for the file path and verifying file existence before processing.
[COMMAND_EXECUTION]: The run-cats-fuzz.sh script uses curl to perform health checks against the local API being tested. These network operations are restricted to the testing environment (localhost or host.docker.internal).

703-technologies-fuzzing-testing