Tracking: v0.2.3 campaign¶
Status: Closed — v0.2.3 shipped with git tag v0.2.3; narrative docs/releases/v0.2.3.md; CHANGELOG [0.2.3]. Next milestone: docs/roadmap/v0.2.4.md · docs/tracking/v0.2.4-campaign.md.
Roadmap (frozen): docs/roadmap/v0.2.3.md · Prior release: docs/releases/v0.2.2.md · Implementation log: docs/implementation-log.md.
Objective (original)¶
Deliver v0.2.3 so that campaign result packs, fingerprint-driven reporting, readable outputs, drift-safe contracts, and honest browser/MCP boundaries are true in code and docs—without expanding unrelated surface area.
Completed items¶
| Area | Evidence |
|---|---|
| Verification surface hardening | docs/workflows/verification.md; just ci vs just validate-artifacts vs just test; README/AGENTS pointers. |
| Drift checks | tests/test_schema_drift_contracts.py; just validate-artifacts; docs/audits/schema-drift-contracts-inventory.md. |
| Campaign result-pack refinement | alwm benchmark campaign pack / pack-check; campaign_result_pack; examples examples/campaign_result_packs/; docs/workflows/benchmark-campaigns.md. |
| Fingerprint interpretation | fingerprint_axis_interpretation (and related) in analysis JSON; comparative + longitudinal Markdown; aligned with group_snapshots_by / longitudinal semantics. |
| Browser reporting realism | Browser evidence in reports where applicable; fixture-backed paths default in CI; docs/architecture/browser.md + docs/audits/capability-classification.md. |
| Pack comparison | alwm benchmark campaign compare-packs; kind campaign_result_pack_comparison; examples/campaign_result_packs/compare_minimal_vs_multi/. |
| Campaign directory comparison | alwm benchmark campaign compare; kind campaign_compare; examples/campaign_compares/ (see CHANGELOG). |
| Report readability | Campaign/comparative/semantic Markdown structure passes (see implementation log entries). |
Release closure (done)¶
[0.2.3]in CHANGELOG.- docs/releases/v0.2.3.md.
README.md/AGENTS.mdLatest shipped updated to v0.2.3; current milestone retargeted to v0.2.4.
Open questions (carry forward to v0.2.4+)¶
- Deduplication: executive vs appendix blocks—editorial preference.
- Fingerprint narrative: template-generated vs hand-tuned copy.
- Redaction policy for packs shared outside the repo.
- Campaign-level semantic-stability thresholds—docs vs harness later.
- Stronger codegen (Schema + Pydantic single source)—post-milestone architecture decision.
(Resolved for v0.2.3: dedicated pack comparison is compare-packs + pack-compare.json.)
Follow-up work (v0.2.4 and beyond)¶
See docs/roadmap/v0.2.4.md and docs/tracking/v0.2.4-campaign.md. Historical candidates (pre–v0.2.4 listing): codegen/drift reduction; deeper diff UX; zip/tar handoff; campaign semantic threshold metadata.
Verification commands (historical)¶
Authoritative list: docs/roadmap/v0.2.3.md — Verification commands.
See also¶
- docs/roadmap/v0.2.0.md — fingerprints + campaigns arc.
- docs/tracking/campaign-orchestration.md — orchestration tracking.
- docs/workflows/benchmark-campaigns.md — campaigns, packs, compare.
- docs/workflows/verification.md — canonical verification matrix.