memory

Multi-perspective code review orchestrator — PR diff analysis with severity-rated findings and LGTM verdict

Phase-based project harness with PRD generation, consensus review, plan validation, multi-mode deliberation, DAG execution, cost forecasting, and decision memory

Web dashboard for .xm project state — visualize builds, probes, solvers, reviews, evals, humble lessons, traces, memory

Strategy orchestration — 18 strategies with self-scoring, --verify auto-validation, and compose presets

Agent output quality evaluation — multi-rubric scoring, diff analysis, reusable judge prompt, strategy benchmarking, and A/B prompt experiments

All-in-one bundle: build + op + agent + solver + eval + humble + cross-plugin quality pipeline

Agent primitives and autonomous behaviors — fan-out, delegate, broadcast, research, solve, consensus, swarm for native multi-agent orchestration

Premise validation — evidence-grade questioning, domain-aware probing, kill bad ideas early, earn the right to build

Multi-agent toolkit for Claude Code. Structured problem solving, strategy orchestration, and agent primitives in a single plugin bundle.

Structured retrospective — reflect on failures together, find root causes, explore alternatives, and grow