The Agent Skills Directory

[SAFE]: The skill defines a defensive security architecture, explicitly prohibiting network connections and non-standard cryptography to maintain a strict local-only data privacy model.- [SAFE]: It implements clear agent-facing guardrails, requiring the AI to halt and notify the user if a proposed change violates key security invariants defined in the repository's philosophy.- [SAFE]: The documentation includes instructions for robust data isolation, zeroization of sensitive memory, and mandatory error sanitization to prevent sensitive data leaks through the user interface or IPC boundaries.- [SAFE]: No suspicious patterns, such as prompt injection, hidden code, or unauthorized data access, were detected in the skill's instructions, frontmatter, or checklists.

security-stance