The Agent Skills Directory

[SAFE]: The skill consists of Markdown instructions designed to guide an AI agent in applying a specific sociological theory. It does not include any scripts, executables, or code blocks that would be run on the host system.
[SAFE]: No network operations or external download patterns (such as curl, wget, or package manager commands) are present in the skill files.
[SAFE]: There are no attempts to access sensitive file paths, environment variables, or hardcoded credentials. The data handled is restricted to user-provided scenarios for academic analysis.
[SAFE]: The instructional constraints, such as the "IRON LAW" in the assumptions section, are domain-specific requirements to ensure the agent adheres to the chosen theoretical framework. These do not constitute prompt injection as they do not attempt to bypass safety filters or disregard system instructions.
[SAFE]: No obfuscation, persistence mechanisms, or privilege escalation techniques were detected.

grad-embeddedness