Dragan Spiridonov 4409699a2b feat(test-gen): enable paid escalation + wire D9 routing-feedback at kernel layer ...

Completes the cheap-first economics: local → repair → escalate the hard tail.

- coordinator: build a claudeRunner from the existing HybridRouter (tier →
  complexity hint) and run the free-tier path with escalate:true. Falls back to
  local-only automatically when no router is wired. patternsUsed now records the
  winning tier.
- plugin (kernel layer): onInitialize constructs a live RoutingFeedbackCollector
  (EMA calibration + auto-escalation, memory-only fallback) when the free tier is
  opted in, and injects it as the coordinator's routingFeedback sink — so each
  cheap-vs-escalated outcome drives the calibrator/escalation tracker/confidence
  (lifts the stuck-at-40% routing metric). Built only when opted in.
- tests: +2 (escalation via router, D9 sink invocation); 113 green; existing
  coordinator (47) and plugin (38) suites unchanged. Type-safe (MessageRole /
  TaskComplexity / RoutingConfig fields all confirmed).

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

2026-06-20 12:43:22 +00:00

..

aqe-mcp-snapshot

feat(metaharness): Darwin-for-QE self-learning lane (A1 scan + D0/D1/D6)

2026-06-20 07:40:05 +00:00

prototype

feat(routing): D7-wire — FreeTierEscalatingExecutor (cheap-first, verify, escalate)

2026-06-20 11:16:48 +00:00

05-cross-pollination-goap-plan.md

docs(metaharness): update cross-pollination plan after MetaHarness pull + ADR-150

2026-06-17 13:12:00 +00:00

06-darwin-qe-self-learning-action-lane.md

feat(test-gen): enable paid escalation + wire D9 routing-feedback at kernel layer

2026-06-20 12:43:22 +00:00

harness-plan.json

feat(metaharness): Darwin-for-QE self-learning lane (A1 scan + D0/D1/D6)

2026-06-20 07:40:05 +00:00

repo-profile.json

feat(metaharness): Darwin-for-QE self-learning lane (A1 scan + D0/D1/D6)

2026-06-20 07:40:05 +00:00