The Agent Skills Directory

[COMMAND_EXECUTION]: The skill involves running local Python scripts included in the package to generate architecture diagrams and dependency reports. These scripts are invoked by the agent to process the user's project files.
[DATA_EXPOSURE]: The included analysis scripts scan the local file system to identify module imports, project structures, and package configurations. This data is used solely to provide architectural insights to the agent and is not exfiltrated to any external destination.
[PROMPT_INJECTION]: The skill has an indirect prompt injection surface because its reports include data extracted from the analyzed project (such as directory names and source code patterns). A malicious project could theoretically include filenames or code comments designed to influence the agent's interpretation of the architecture, though the impact is localized to the analysis output.

senior-architect