next-issue@16.12.1
Autonomous issue harness driven by the Claude Agent SDK
AI security review
scanned · by lpm-firewall-aiDefault CLI startup writes or refreshes the package’s own Claude skills in the project, which is an agent extension lifecycle risk. The skill contents match the package’s issue workflow, and the inspected code does not show broader agent setting changes or unrelated behavior, so this warrants a warning rather than a block.