The Agent Skills Directory

[SAFE]: The skill is designed to read common project configuration and dependency files (e.g., package.json, Gemfile, tsconfig.json) to extract facts about the technology stack.
[DATA_EXFILTRATION]: Although the skill accesses various project files, it only stores the extracted facts using the agent's internal memory tools (memory.store_extraction and memory.recall). No network operations or external data transfer commands were detected.
[PROMPT_INJECTION]: The instructions are strictly focused on project analysis and do not contain any attempts to override safety filters or hijack agent behavior.
[COMMAND_EXECUTION]: The skill does not execute arbitrary shell commands or scripts. It primarily uses file-reading capabilities to gather information.

project-analysis