autonomous-loops

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.85). The skill's workflow explicitly instructs fetching and executing a remote install script from raw.githubusercontent.com and has the agent read CI logs via gh run view (public GitHub content) as part of its autonomous fix-and-merge loop, meaning untrusted, user-generated third-party content can be ingested and directly influence agent actions.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill instructs users to run a remote install script that will be executed locally via: curl -fsSL https://raw.githubusercontent.com/AnandChowdhary/continuous-claude/HEAD/install.sh | bash, which fetches and executes remote code at runtime and is presented as the installation step for the Continuous Claude workflow.