feat: implement issue #1898 — [Phase 2] Swap the deep tier to claude-opus-5-5 in engine.sh (model, chain, label, effort)

Merged
DO
1mo ago
+166-64

Problem

[Phase 2] Swap the deep tier to claude-opus-5-5 in engine.sh (model, chain, label, effort)

From the issue: As a review-engine maintainer, I want point ENGINE_DEEP_MODEL and the deep fallback chain at claude-opus-5-5 and update the engine label, so that the highest-cost agentic tier (deep review, dev-lead fix-issue, persona advisory) runs on the cheaper, higher-benchmark model.

Risk

Low — changes automation shell logic under scripts/, covered by shellcheck (--severity=warning) and the bats suite.

Test plan

Tests added/updated: tests/dev-lead/unit/test_engine_chain.bats,tests/dev-lead/unit/test_engine_mcp.bats tests/dev-lead/unit/test_engine_new_models.bats. Verification: bash scripts/dev-lead-lint.sh (shellcheck --severity=warning) ran pre-commit; the bats suite runs in CI.

Rollback

Revert this PR. No non-revertible side effects (no tags, migrations, or external state).

Monitoring

This PR's Lint (shellcheck) and bats checks show pass/fail; watch subsequent dev-lead / pr-review runs for behavioral regressions.

Closes #1898

Summary by CodeRabbit

  • Improvements
    • Deep-tier Claude runs now use Claude Opus 5.5 with high reasoning effort. If that model is rate-limited, the system can fall back to Claude Sonnet 5.0.
    • Advisory runs now explicitly use Claude Opus 5.5 with high reasoning effort.
    • Other Claude tiers continue without a reasoning-effort override.