The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute various CLI tools, including python, uv, ruff, and git. These tools are used for running verification scripts, performing linting, and managing the local repository state to establish baselines for security findings.- [DYNAMIC_EXECUTION]: The script redaction_check.py dynamically computes the repository root path using pathlib and modifies the sys.path at runtime. This allows it to import the project's actual redaction patterns from utils.logging, ensuring that the verification tests accurately reflect the production environment.- [INDIRECT_PROMPT_INJECTION]: The skill defines a surface for processing external data (log shapes and potential secrets) to verify filtering logic. While it handles untrusted input, the operations are limited to string substitution and regular expression matching for testing purposes, which is the primary function of the utility.

verify