v10.2.0

What's Changed

• release: v10.2.0 — simulator agent and skill-first dispatch by @nullhack in #192

Full Changelog: v10.1.0...v10.2.0

v10.1.0

What's Changed

• fix(release): align with v9 model by @nullhack in #189
• fix(release): rename workflow to match PyPI trusted publisher by @nullhack in #190
• feat(pipeline): refresh state, cache bootstrapping, rework triggers, branch discipline (v10.1.0) by @nullhack in #191

Full Changelog: v10.0.0...v10.1.0

v10.0.0 — clean-slate methodology rebuild

What changed

A clean-slate rebuild of the methodology layer, with five dogfood fixes that tighten the staged-contract pipeline, a test-driven narrative reset, and a release workflow with trusted PyPI publishing.

Pipeline fixes (dogfood)

• model-data-schema state (plan entry) — the Data Architect authors data-model.md as a binding input to every downstream plan state. Workload is classified OLTP / OLAP / hybrid; every field is traced to a stakeholder need. No more "we built the contract surface, then discovered the schema was wrong."
• simulate-contracts is now a compiler, not a ceremony. It walks the contract surface hop-by-hop, journals one observation per hop, runs a spec-diff (catches what two implementations could both satisfy), and surfaces build-implied gaps (effect-without-mechanism, side-effect-not-observed, model-disagreement) before any source .py exists.
• review-test-stubs gate adds a vacuous-assertion sweep (five named smells: hasattr-only, no-op helper, lower-bound-only, constant-satisfiable, tautology) and an inverted-traceability (scope-minimal) check — every test maps to a finding, or it goes.
• Simplicity is a traced requirement. Each interview funnel level asks a simplicity question (smallest surface, speculative vs grounded, load-bearing vs collapsible); consolidate-interview drops speculative items before they become contracts.
• Replay is unambiguous. HTTP adapters replay through vcr.use_cassette(); CLI subprocess tests use pytester; library-boundary adapters (ddgs, primp) use monkeypatch. The skill layer names the kind and the fixture.
• Journal template at .templates/cache/journal.md.template — first-pass emptiness is expected; three sections match the three writer triggers.

Template & narrative

• Test-driven narrative. README, About, and pyproject.toml description reframed from spec-driven to test-driven ("Tests are the contract").
• conftest.py moved to tests/ — the pending-marker hook is test-only; root placement was a template-portability holdover.
• Branding assets (docs/assets/{banner,logo}.svg) pulled back into the template.
• Explicit [build-system] (setuptools≥68) per PEP 517; [tool.setuptools] packages = [] (template ships no package; wheel is metadata-only for PyPI discoverability).
• release.yml workflow — on release published, uv build + trusted-publishing to PyPI (OIDC, no token) + upload wheel/sdist to the release assets.

Internal

• Plan flow v1.6.0 (new model-data-schema entry state).
• New skill: model-data-schema. New knowledge: architecture/data-modeling.md, methodology/simplicity-discipline.md.
• Gate evidence keys: vacuous-assertion-free, scope-minimal, spec-diff-clean.

Full diff: v9.4.0...v10.0.0

v9.4.0: Branch Lifecycle Discipline

Changed

• merge-local/SKILL.md: sync local main with remote before merge, reset dev to origin/main to prevent history divergence, delete feature branch after squash-merge
• create-pr/SKILL.md: after PR merged, sync local main and reset dev to origin/main for clean next cycle
• structure-feature/SKILL.md: pull remote main before creating feature branch from current main
• git-conventions.md: added Branch Lifecycle concept (short-lived feature branches, only main/dev persist), updated Conflict Prevention and Administrative PR sections with post-merge sync discipline

v9.3.0: V10 Agentic Reform

Changed

• AGENTS.md: full rewrite — caveman-style throughout, Main Directive at top, Golden Rules tightened, feature lifecycle phases documented, runtime artifact resolution convention, review loop relay protocol, loop exit detection in anchor, multi-skill ordering explicit, beehave violations = test failures, convention boundary for design-phase states
• refactor/SKILL.md: added beehave check before task test-fast (step 9), added loop exit detection via beehave status --json (step 10)
• write-test/SKILL.md: added no-docstrings rule for test functions with self-heal ruff instruction
• review-gate/SKILL.md: added no-docstrings enforcement in structure review (step 8)
• structure-feature/SKILL.md: clarified out: git_branch as Runtime artifact meaning branch creation
• todo-anchor-protocol.md: synced notation to native status fields, added loop resolution and review relay to anchor checklist
• test-design.md: added comment/docstring noise pattern to Spec Value Fidelity
• test-stubs.md: documented auto-generation on every pytest, stale stub accumulation, beehave clean --force usage
• skill-design/principles.md: added imperative language standard for skill writing
• pyproject.toml: added "D" to per-file-ignores for tests/** (no docstrings required in tests)

Simulation-Driven

• Walked full 24-state define-flow → develop-flow path for "Add Task" feature
• Identified and resolved 5 critical pain points: runtime artifact ambiguity, feature file lifecycle, multi-skill ordering, review feedback survival, loop exit detection

v9.2.1 — Steadfast Hephaestus

Steadfast Hephaestus — 2026-05-22

Upgrade flowr 1.1.0 → 1.2.0 (Whole Rye) to fix flowr serve crash from missing static frontend assets in the wheel.

Fixed

• flowr serve crash: upgraded flowr from 1.1.0 to 1.2.0 (Whole Rye) which includes the missing static frontend assets in the wheel. The flowr serve command now finds index.html, css/, and js/ and starts the visualization server without errors.
• Dropped html extra: flowr[html,viz] changed to flowr[viz] — the html extra never existed in flowr and caused a resolution warning.

Changed

• flowr: >=1.1.0 → >=1.2.0
• flowr[viz]: >=1.0.0 → >=1.2.0

v9.2.0 — consistency audit, feature_title convention, 7-principle review

Changes

Convention fixes

• feature_id → feature_title: Unified across 5 flow YAML files, 4 skills, 1 knowledge file, AGENTS.md, CHANGELOG.md. Template rename: <feature_name> → <feature_title>.
• flowr 1.0.0 → 1.1.0: pyproject.toml and uv.lock.

Design principle review (7 checks)

• review-gate: step 5 split into 5a–5g — one sub-step per design principle (YAGNI, KISS, DRY, ObjCal, Smells, SOLID, Patterns), each matching tdd-cycle exit conditions with dedicated knowledge references.
• refactor: explicit knowledge references added for all 7 principles at detection point.

beehave status integration

• verify-traceability: beehave status --json now mandatory gate — stage ok skips beehave check.
• review-gate tier 2: beehave status --json as pre-check before beehave check.

Flow YAML fixes

• Added params: [feature_title] to define-flow.yaml
• Removed unreachable cancelled exit from develop-flow.yaml
• Removed stale needs-reinterview exit from define-flow.yaml

Agent and skill ownership

• SE agent: explicit ownership — fix existing issues regardless of origin, never noqa, never pyproject.toml
• implement-minimum: fix existing test/beehave failures in touched files
• polish-code: fix all violations codebase-wide
• write-test: anti-noise rules for Spec Value Fidelity

Templates

• New test stub template `.templates/tests/features/<rule_slug>_test.py.template"
• Template rename: <feature_name> → <feature_title>

Cleanup

• Removed stale @id references from AGENTS.md, README.md, docs/audit/
• Updated test-stubs.md to use <feature_title> convention

v9.1.0 — beehave 1.0.0 knowledge integration

Knowledge

• test-stubs.md: Full rewrite incorporating beehave 1.0.0 features — beehave status (6 development stages, --json, --include-unmapped), beehave list/list -v, beehave clean, collision detection, feature stem = title slug requirement
• source-stubs.md: Updated creation order to include beehave status
• gherkin.md: Updated Related link to reference beehave status

Skills

• select-feature: Replaced manual checking with beehave status --json for project-wide overview
• accept-feature, verify-traceability: Added beehave status --json pre-check
• review-gate: Added test-stubs to available knowledge

Project

• Added [dependency-groups] with flowr>=1.0.0 in dev (PEP 735)

v9.0.1

What's Changed

• release: bump version 9.0.0 → 9.0.1 (derive delivery order from context map) by @nullhack in #162

Changed

• Derived delivery order: Removed static Delivery Order section from product_definition.md. Feature selection now derives order from context map dependency count (lowest first), ties broken by WSJF.
• Research template restructured: Moved to domain/subdomain hierarchy.
• Template cleanup: Simplified domain_spec, glossary, and product_definition templates.

Full Changelog: v9.0.0+20260518...v9.0.1

v9.0.0+20260518

What's Changed

• release: bump version 8.6.0 → 9.0.0 (Adversarial Audit) by @nullhack in #160

Full Changelog: v8.6.0+20260515...v9.0.0+20260518