The Agent Skills Directory

[SAFE]: Robust verification process. The skill requires the agent to verify all sub-agent claims against the original source files, effectively mitigating risks from potential misinformation or malicious instructions within analyzed documents.
[SAFE]: Restricted execution environment. All operations are limited to reading and editing local project markdown files, with no requests for network access, external tool invocation, or sensitive credential retrieval.
[SAFE]: Transparent logic. The skill contains no obfuscated instructions, encoded URLs, or hidden dependencies, and its behavior aligns strictly with its stated purpose of document auditing.

review-plan