Stage 2: Compound -- KARL Phase 4+ Unified Architecture

We have a trajectory recording system (110 records), a reward engine (3-signal composite), a shadow vector router (10% cache hit rate), and a training pipeline that produced one adapter (KARL v2, loss 1.843, gemma-3-1b-4bit) from 35 SFT examples. The adapter exists but has never been evaluated for actual routing or planning quality. The finetune daemon on Mac5 is down. The promotion gate says the shadow router is not ready. **The honest assessment**: We have a data collection system that works and a training pipeline that runs. We do not yet have evidence that the trained model improves anything. The gap between "model trained" and "model useful" is the gap this compound must close. Before any data or algorithm work, the training infrastructure must be reliable and observable. **Actions**: 1. SSH to Mac5, restart finetune daemon, create LaunchAgent `com.openclaw.finetune-daemon.plist` with auto-restart 2. Increase training hyperparameters: seq_len 256->512, LoRA rank 8->16, layers 4->8, batch_size 1->2 3. Create `[home-path]` with three evaluation functions: - `evaluate_routing_accuracy()`: Given test prompts, does the model predict the correct skill? - `evaluate_planning_quality()`: Given tasks, does the model generate tool plans that match reference? - `compare_adapters()`: Head-to-head comparison between two adapter versions 4. Build a held-out evaluation set: 20 real prompts with known-correct skills and tool plans, manually curated from the 34 high-reward trajectories. This set is NEVER used for training. **Validation gate**: Finetune daemon responds on :9200. Evaluator runs successfully on KARL v2. Baseline metrics recorded.