{"id":"andyzengmath-quantum-loop","name":"quantum-loop","af_score":53.0,"security_score":12.5,"reliability_score":31.2,"what_it_does":"Quantum-Loop is a Claude Code plugin plus scripts that implements a spec-driven autonomous coding loop: it turns a one-line feature description into structured PRD/spec artifacts, builds an execution dependency DAG, runs TDD/quality checks with two-stage review gates, enforces “fresh verification evidence” (Iron Law), and iterates with retry/failure logging until stories pass and then performs cross-story integration before committing changes.","best_when":"You want a disciplined, autonomous coding workflow that minimizes spec drift and “looks good” claims by using evidence-based verification and gated reviews—especially for multi-story changes with dependencies.","avoid_when":"You need a simple one-shot code generator or you can’t permit running multiple background agents/worktrees and committing to a local repo.","last_evaluated":"2026-03-30T15:37:11.266342+00:00","has_mcp":false,"has_api":false,"auth_methods":["Local plugin / CLI usage inside Claude Code"],"has_free_tier":false,"known_gotchas":["Parallel worktrees may still cause conflicts if story file-conflict filtering misses edge cases; merges/retries depend on the orchestrator’s filtering and git behavior.","Agents must commit before signaling; if an agent crashes or times out, uncommitted work in its worktree can be lost per README.","Python isolation is described as PYTHONPATH injection to avoid editable-install races; if a project relies on editable installs, behavior may differ.","Parallel agents share a single Python interpreter; concurrency-sensitive test/build steps could still interfere depending on project tooling.","Timeout/stale detection defaults (e.g., 15 min execution, 20 min staleThreshold) may cause extra retries for slow CI-like tasks."],"error_quality":0.0}