The Agent Skills Directory

[SAFE]: The skill is entirely instructional and theoretical, defining a protocol for objective information processing. It contains no executable code, shell commands, or network operations.
[PROMPT_INJECTION]: The instructions to "clear accumulated narrative" and "break the chain of conditioned reactions" are presented as methods for enhancing analytical clarity and objectivity. These do not target system safety filters or attempt to override core agent instructions.
[DATA_EXFILTRATION]: There are no patterns involving network access or the reading of sensitive system files. The information processing described occurs entirely within the existing conversation context.
[REMOTE_CODE_EXECUTION]: The skill does not perform any package installations or remote script executions. References to external tools are conceptual and pertain to potential integration with other skills rather than active code execution.

vipassana-llm