The Agent Skills Directory

[SAFE]: The skill uses a limited set of tools (Read, Grep, Glob) and does not request network access or execution permissions, minimizing the risk of data exfiltration or system compromise.
[SAFE]: The skill includes a dedicated security notice instructing the agent to treat input data as untrusted, mitigating the risk of indirect prompt injection from configuration files.
[SAFE]: The core logic is focused on architectural analysis and reporting based on established security frameworks (NIST SP 800-207 and CIS Controls v8), with no evidence of malicious intent or obfuscation.

segmentation