Phase 4.5 intelligence layer

Full end-to-end lifecycle test — seed agent, generate sessions with behavior shift, verify fingerprint/drift/regression/promotion flow — plus multi-page Playwright scenario.

Milestone 4.5.D.2 — E2E Playwright Coverage

Status: Planned
Goal: Track D — Exit Gate
Phase: 4.5 — Intelligence Layer
Estimated effort: 2 days
Track: Track D — Exit Gate


Why This Milestone Exists

The highest-value Playwright scenario in the redesigned roadmap — the first place three previously-separate subsystems (fingerprinting, drift, directive regression) are proven to work as one operator-facing workflow through the dashboard.


E2E Scenarios

API-level (make e2e-smoke-p4.5): Seed agent → 30 synthetic sessions with mid-stream behavior shift → verify fingerprint → drift alert → regression run → promotion blocked pending review.

Playwright: Operator opens drift alert → clicks "run regression from here" → watches live pass/fail stream → approves → promotes with gradual rollout → observes rollout progress advancing → manually pauses → resumes.


Orientation (indicative)

Named paths, package layouts, libraries, schemas, env vars, and commands anywhere on this page are rough sketches for orientation — inspiration and a baseline, not a required change list.

During implementation, expect to:

  • open the live tree and follow existing patterns before inventing new ones
  • research current constraints (latency, tenancy, deploy shape, libraries) more deeply than this page can
  • advance the design beyond the sketch where measurement or code reality says so
  • land work in different filenames, merged packages, deferred docs, or new surfaces when the situation calls for it

Prefer outcomes over matching any particular file tree or command sequence.

Areas that may be involved (situational — not a checklist):

  • Evaluation / benchmarks
  • Fingerprinting / drift
  • Directive regression / promotion

Success signals

Outcome-oriented signals that the milestone is in good shape. Exact filenames, package layouts, and commands may differ from any sketches above.

  • API-level E2E exits 0 with full lifecycle verified
  • Playwright flow covers drift-alert → regression → promote → rollout → pause → resume
  • Both scenarios run in CI (smoke profile)

Prerequisites

  • Milestone 4.5.D.1 (Cross-track integration suite passing)
  • Milestone 4.5.C.4 (Dashboard integration)
Edit on GitHub

Last updated on

On this page

0%