The Agent Skills Directory

[SAFE]: The skill is primarily composed of markdown documentation, checklists, and templates designed to guide the code review process. It does not contain any executable scripts or automation that performs sensitive operations.\n- [SAFE]: The code snippets provided in SKILL.md and the examples directory are for illustrative purposes only, showing how to integrate with the Anthropic SDK for automated reviews. These snippets are contained within markdown code blocks and are not invoked by the skill.\n- [SAFE]: Analysis for obfuscation (Base64, zero-width characters, homoglyphs) found no malicious use. Zero-width spaces detected in the walkthrough example are used correctly for markdown formatting of nested code blocks.\n- [SAFE]: The skill does not access sensitive files, perform unauthorized network requests, or attempt to persist access. It provides defensive security guidance (e.g., checking for SQL injection in code under review) without introducing vulnerabilities itself.

code-review-playbook