The Agent Skills Directory

[SAFE]: The skill is entirely composed of Markdown and YAML instructions. It does not contain any executable scripts, tool definitions, or subprocess commands.\n- [SAFE]: No malicious patterns, such as prompt injection, credential exfiltration, or obfuscation techniques, were identified in the skill files.\n- [SAFE]: The skill references external repositories for attribution purposes and provides installation instructions for the user. These references are documented neutrally and are consistent with the vendor's own repository structure.\n- [SAFE]: The skill ingests user-provided code to perform audits or generate comments, creating an indirect prompt injection surface. Evidence chain: 1. Ingestion points: User-provided code snippets or file contents processed via prompt instructions. 2. Boundary markers: Absent (the skill provides templates but does not define strict delimiters for user code). 3. Capability inventory: None (the skill lacks access to the file system, network, or shell commands). 4. Sanitization: Absent. The lack of system capabilities effectively mitigates the risk of exploitation.

code-comments