@miller-tech/uap 1.31.0 → 1.32.0

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
package/README.md CHANGED
@@ -91,7 +91,7 @@ uap setup -p all
91
91
  | Browser | 1 module | Stealth web automation via CloakBrowser (Playwright drop-in) |
92
92
  | MCP Router | 11 modules | 2-tool meta-router + expert-consultation registry (98% token savings) |
93
93
  | Models | 10 modules | Multi-model routing, planning, execution, validation, 13 model profiles |
94
- | Delivery Harness | 11 modules | `uap deliver`: convergence loop, best-of-N explorer, critic, practice recall, escalation, ideation seeds, HALO tracing, coordination + deploy queueing |
94
+ | Delivery Harness | 12 modules | `uap deliver`: convergence loop, best-of-N explorer, critic, practice recall, escalation, ideation seeds, HALO tracing, coordination + deploy queueing |
95
95
  | Patterns | 23 patterns | Battle-tested workflows from Terminal-Bench 2.0 |
96
96
  | Droids | 30 experts | Full SDLC expert stack: strategy, design, build, review, release, ops ([reference](docs/reference/EXPERT_DROIDS.md)) |
97
97
  | Expert Orchestrator | 1 module | Adaptive droid-chain selection across plan→design→implement→review→release |
@@ -373,8 +373,9 @@ the model's say-so.
373
373
  8. **Coordination** (`--coordinate`) — registers the run with the multi-agent coordination layer (`uap agent`): announces work on the project, warns about overlapping agents, heartbeats every turn, completes/deregisters on exit.
374
374
  9. **Deploy batching** (`--deploy`) — on success, queues a commit of the applied files into the deploy batcher; execute with `uap deploy flush`.
375
375
  10. **`--optimize`** — one switch for every convergence aid: 4 candidates/turn + critic + practices + escalation + ideation + HALO + coordination (deploy stays explicit).
376
+ 11. **Dynamic optimization (default)** — every instruction is classified for complexity (simple / moderate / complex); non-trivial requests automatically get the aids that improve outcomes (moderate → exploration ×3 + critic + practices + HALO + coordination; complex → the full `--optimize` stack). Any explicit aid flag, `--no-auto`, or `UAP_DELIVER_AUTO=0` disables auto mode. Deploy queueing is never auto-enabled.
376
377
 
377
- ### Components (11 modules)
378
+ ### Components (12 modules)
378
379
 
379
380
  | Component | File | Purpose |
380
381
  | ----------------- | ------------------------------------- | ----------------------------------------------------------------- |
@@ -389,6 +390,7 @@ the model's say-so.
389
390
  | Ideation Seeder | `src/delivery/ideation.ts` | Divergent strategy seeds (generated or from curated ideas) |
390
391
  | HALO Tracer | `src/delivery/halo-trace.ts` | Run/turn spans for `uap harness analyze` |
391
392
  | Run Coordinator | `src/delivery/run-coordinator.ts` | `uap agent` registration/heartbeat + `uap deploy` commit queueing |
393
+ | Auto-Optimizer | `src/delivery/auto-optimizer.ts` | Complexity-classified dynamic activation of convergence aids |
392
394
 
393
395
  The model is reached through an OpenAI-compatible client
394
396
  (`src/models/openai-compat-client.ts`) — the local inference gateway,
@@ -436,6 +438,7 @@ uap deliver "..." --ideate --candidates 4 --deploy
436
438
  | `--coordinate` | Register with `uap agent`: announce, heartbeat, overlap detection |
437
439
  | `--deploy` | On success, queue a commit into the deploy batcher (`uap deploy`) |
438
440
  | `--optimize` | Enable every convergence aid (deploy excluded) |
441
+ | `--no-auto` | Disable dynamic optimization (auto-classified aids are the default) |
439
442
  | `--endpoint <url>` | Override the model endpoint (OpenAI-compatible `/v1`) |
440
443
  | `--dry-run` / `--json` | Show the plan only / emit machine-readable result |
441
444