The Agent Skills Directory

[SAFE]: The skill automates the setup of react-doctor, a legitimate static analysis tool for React codebases, and configures it to run as a hook.
[SAFE]: The bash script scripts/react-doctor-stop.sh demonstrates secure coding practices by using jq -Rs . to properly escape raw tool output before returning it as a JSON decision. This prevents potential indirect prompt injection that could occur if the tool output contained instructions intended for the agent.
[SAFE]: No unauthorized network requests, credential harvesting, or obfuscation techniques were found in the instructions or scripts.

setup-react-doctor