@primitive.ai/prim@0.1.0-alpha.118
CLI for Primitive's decision graph — passive decision capture, conflict gate, and team presence
AI security review
scanned · by lpm-firewall-aiThe session-start hook injects coercive instructions that tell Claude Code to block unrelated task work and keep pressing for a repository connection even after the user declines. The connection command requires approval, so this is suspicious agent capability abuse rather than an automatic malicious account action.