The Agent Skills Directory

[SAFE]: The skill provides structured guidance for the creation and maintenance of agent instruction files and does not contain any malicious patterns or bypass attempts.\n- [SAFE]: Explicit operational gates are defined, ensuring that the agent only creates, edits, or appends files when explicitly prompted or permitted by the user, which prevents unauthorized filesystem operations.\n- [SAFE]: The prompt discipline rules encourage affirmative phrasing and signal density, which are best practices for reliable agent behavior and do not involve any safety-filter bypasses.\n- [SAFE]: No evidence of data exfiltration, credential exposure, or unauthorized network activity was detected across the skill's instructions or its reference files.\n- [SAFE]: Reference files include security-conscious examples, such as checklists to ensure no secrets are included in repository diffs.

using-agentsmd