The Agent Skills Directory

[COMMAND_EXECUTION]: The skill facilitates the execution of local scripts and standard version control tools for the purpose of regression testing. * Evidence: The workflow utilizes python tests/uat/run_uat.py to execute dynamically designed test scenarios. * Evidence: Git commands such as git fetch, git checkout, and git pull are used to manage branches for behavior comparison.
[PROMPT_INJECTION]: The skill processes user-provided descriptions to generate test scenarios, creating an interface for indirect prompt injection. * Ingestion points: User instructions provided via $ARGUMENTS are used to generate the setup, test, and teardown prompts within the scenario JSON. * Boundary markers: Not explicitly defined in the prompt generation instructions. * Capability inventory: Access to Bash, Read, and Write tools, along with the ability to execute the local Python test runner. * Sanitization: No explicit sanitization or filtering of the user-provided descriptions is mentioned before interpolation.
[SAFE]: The skill does not exhibit any patterns of obfuscation, unauthorized network operations, or credential harvesting. * Evidence: Network activity is limited to standard git operations on the local repository. * Evidence: All scripts and tools referenced are local to the development environment.

bat-adhoc