The Agent Skills Directory

[SAFE]: The skill consists entirely of instructional prompting designed to guide the agent's internal reasoning and problem-solving methodology. It does not contain any executable scripts or active code components.- [NO_CODE]: There are no shell commands, system calls, or scripts included in the skill files. The methodology is expressed in natural language without any platform-specific automation or tool execution.- [EXTERNAL_DOWNLOADS]: The skill does not attempt to download external packages, scripts, or configuration files. It specifically notes that no APIs or authentication are required.- [DATA_EXFILTRATION]: No network-enabled commands or sensitive file access patterns were detected. The skill operates entirely within the model's existing context window.- [PROMPT_INJECTION]: The instructions focus on improving logical consistency and self-correction during reasoning. There are no patterns suggesting attempts to bypass safety filters or override system constraints.

sequential-thinking