The Agent Skills Directory

[SAFE]: The skill is strictly documentation-based, providing policy and procedural guidance for application security management. No active code or malicious functionality was identified.\n- [NO_CODE]: No scripts (Python, JavaScript, Shell) are included in the skill files. All content is educational or instructional markdown.\n- [METADATA_POISONING]: Frontmatter metadata is accurate and describes the skill's purpose without incorporating deceptive or malicious instructions.\n- [INDIRECT_PROMPT_INJECTION]: The skill defines a process for reviewing external data. 1. Ingestion points: Source code submitted for review against references/secure-code-review-checklist.md. 2. Boundary markers: Absent. 3. Capability inventory: No tool usage, file system access, or network capabilities detected. 4. Sanitization: Absent. Given the lack of executable capabilities, this represents a safe advisory surface.

application-security