The Agent Skills Directory

[SAFE]: The skill provides documentation templates and organizational guidelines for experiment logging. It contains no executable code or scripts within its files.\n- [DATA_EXFILTRATION]: No network activity or unauthorized data access patterns were found. The skill defines structures for local data storage within the project environment (e.g., experiments/, analysis_outputs/).\n- [COMMAND_EXECUTION]: While the skill mentions recording and running 'commands' for experiments, it does not include any malicious scripts or direct execution of untrusted code. It facilitates the logging of a researcher's own execution steps.\n- [PROMPT_INJECTION]: No instructions to bypass safety filters or ignore system rules were detected. The skill maintains clear instructional boundaries.

experiment-logbook