react-doctor

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). The skill deliberately instructs the agent to fetch and follow a remote "playbook" (and per-rule prompts) that can edit the working tree directly without commits or PRs, creating a high-risk supply-chain/backdoor vector for remote code execution or unauthorized repository modification.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 1.00). The skill fetches and re-fetches remote prompt files at runtime which directly control agent instructions (curl to https://www.react.doctor/prompts/react-doctor-agent.md and on-demand fetches from https://www.react.doctor/prompts/rules//.md), so the agent's behavior is driven by externally hosted content.