The Agent Skills Directory

[SAFE]: The skill operates purely as a text-processing utility, refining user input through pre-defined templates. It does not request or use any tools that would allow for data exfiltration or system modification.
[SAFE]: The instructions (e.g., 'anti-sycophancy' clauses) are explicitly designed to improve the truthfulness and robustness of AI interactions, which is a defensive security posture.
[SAFE]: No obfuscation, malicious metadata, or external dependencies were found. The skill uses a local file ('references/patterns.md') for specialized templates.

improve-prompt