The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute local commands using uv run python sct.py. These commands are used for legitimate configuration management tasks, such as previewing sizing constraints (sizing preview), resolving constraints to instances (sizing resolve), and viewing the instance catalog (sizing catalog).
[INDIRECT_PROMPT_INJECTION]: The skill processes external Jenkinsfiles and YAML configuration files to extract test settings. While this represents a surface for indirect prompt injection (where malicious instructions could be embedded in the configuration files), the risk is minimized as the skill primarily uses these files for structured data extraction and passes them to a specific local tool (sct.py).
Ingestion points: Jenkins pipeline files (.jenkinsfile) and test-case configuration files (.yaml).
Boundary markers: None explicitly mentioned in the prompts, though the tool sct.py likely handles parsing.
Capability inventory: File reading and execution of the sct.py script via uv.
Sanitization: The instructions rely on the parse_jenkinsfile() function within the local tool to handle input.

migrate-to-sizing