v8.1.0+20260502 — Disciplined Hermes

Disciplined Hermes

Flow discipline enforced — sessions, branches, and guards keep every step accountable.

Added

• Git branch discipline: Every state in all 12 flow YAML files now declares git: main or git: feature in its attrs. Agents must work on the declared branch and never switch mid-state. Golden Rule #7 added to AGENTS.md.
• Committed-to-main guards: Project-phase exit transitions (discovery, architecture, planning, branding, setup) now require committed_to_main_locally: ==verified evidence before advancing. Ensures artifacts are persisted before moving to the next phase.
• flowr 0.4.0 adoption: Upgraded dependency from >=0.3 to >=0.4. Added [tool.flowr] config section to pyproject.toml (flows_dir, sessions_dir, default_flow, default_session).
• Session management: .flowr/sessions/ directory for persisting workflow progress. AGENTS.md updated with session init, --session flag usage, and session-based workflow pattern.
• Flow params: Feature-scoped flows (planning, development, delivery, tdd-cycle, review-gate, feature-development) now declare params: [feature_name] for session parameter tracking.
• Flowr commands: AGENTS.md and knowledge files updated with session commands (init, show, set-state, list), --session flag on check/next/transition, config command, short flow name resolution, and --evidence-json flag.
• flowr-spec.md knowledge: Session model (flow, state, name, stack, params), configuration resolution, MUST/SHOULD severity levels on validation rules, atomic session writes principle.
• flowr-operations.md knowledge: Session commands, session-based workflow pattern, configuration section, evidence syntax with condition operators.
• Research note: docs/research/software-engineering/process/nullhack_flowr_0.4.0.md documenting flowr 0.4.0 analysis.

Changed

• Flow version bumps: main-flow 7→8, discovery-flow 3→4, architecture-flow 4→5, planning-flow 4→5, feature-development-flow 6→7, development-flow 4→5, delivery-flow 4→5, tdd-cycle-flow 2→3, review-gate-flow 2→3, post-mortem-flow 2→3, branding-flow 2→3, setup-project-flow 2→3.

v8.0.0+20260501 — Foundational Athena

Foundational Athena

From zero to hero — production-ready Python, without the ceremony.

Highlights

• Flow-based delivery system: YAML-driven state machines (.flowr/flows/) replace ad-hoc workflow scripts. Flowr commands (check, next, transition) provide structured session management.
• Research library: Organized research notes and knowledge files under .opencode/knowledge/ with progressive disclosure (#key-takeaways, #concepts, full).
• In/out artifact contracts: Flow states declare explicit in/out attrs — every step reads and writes only what it owns.
• Branding & visual identity: Greek temple logo, parchment/gold palette, and adjective-greek-figure release naming convention.

Changes

• feat: add Boehm 1981 research note, update release workflows, restructure docs index (#128)
• fix: remove all v7 files not present in v8.0.0 (#127)
• feat: v8.0.0 — flow-based delivery system with research library, in/out attrs, and real flowr commands
• refactor: consolidate .flowception/ into .flowr/, add flows/ and sessions/ subdirs
• feat: add flowr dev dependency, add validate-flows task
• fix: align .flowr/ YAML with flowr spec
• docs: add product-definition.md, trim README to landing page, add doc cards
• refactor: remove pytest-beehive dependency, reimplement assign_ids locally
• fix: move ## Changes section into feature description area
• refactor: remove discovery.md, embed per-session changes in .feature files
• refactor: update all references from docs/flows/ to .flowr/
• refactor: relocate flow YAML definitions from docs/flows/ to .flowr/
• refactor(workflow): migrate to YAML flow definitions and knowledge system (#124)
• refactor(workflow): adopt flow definition v1 spec
• refactor(workflow): migrate to YAML flow definitions and knowledge system
• fix(ci): install package in editable mode for PyPI publish quality gate (#115)

v7.2.20260423 - Methodical Theseus

v7.2.20260423 - Methodical Theseus

"Methodical as Theseus threading the labyrinth — every file accounted for, every duplicate purged."

Changelog

Added

• scripts/: 10 validation and automation scripts (assign_ids, check_adrs, check_commit_messages, check_feature_file, check_oc, check_stubs, check_version, check_work_md, detect_state, score_features, update_index_html) (#113)
• docs/adr/ADR-2026-04-22-cli-parser-library.md: decision record for argparse choice (#113)
• .opencode/skills/architect/adr.md.template: ADR template for Step 2 (#113)
• docs/features/completed/cli-entrypoint.feature: completed CLI entrypoint feature (#113)

Changed

• Mermaid → markdown tables: C4 diagrams converted; British English adopted (#113)
• ADR format restructured with Interview Q&A table and stakeholder validation gate (#113)
• Skills, agents, CI workflows, system.md, glossary.md, index.html updated (#113)
• Version references updated from 7.1.20260422 to 7.2.20260423

Removed

• Duplicate template files consolidated under architect/ (#113)
• Stale tests/unit/app_test.py and display-version.feature removed (#113)
• tests/conftest.py manual marker hook removed (#113)

Summary

Consolidation release syncing all uncommitted work from the backup branch — Mermaid-to-markdown conversions, British English adoption, ADR template restructuring, and duplicate file cleanup. Named for Theseus, who methodically retraced every step through the labyrinth.

---

SHA: fac22eb

v7.2.20260422 - Resolute Hermes

v7.2.20260422 - Resolute Hermes

"The herald delivers the interface."

Changelog

Added

• feat: CLI entrypoint — python -m app --help and python -m app --version; zero new dependencies (#109)
• feat: ADR-2026-04-22-cli-parser-library — use argparse (stdlib) over click/typer (#109)
• feat: ADR-2026-04-22-version-source — read version from importlib.metadata at runtime (#109)
• feat: redesigned docs/index.html — branding palette, doc cards, tabbed features, ADR list (#109)

Fixed

• fix: pyright type errors in scripts/check_oc.py and scripts/detect_state.py (#109)
• fix: CI test job now uses venv-scoped uv for importlib.metadata to resolve package (#109)

Changed

• chore: workflow skills updated — Actions/Entities terminology, WORK.md format simplified (#109)

Summary

Resolute Hermes delivers the first working CLI skeleton — a minimal --help and --version entrypoint that demonstrates the full five-step delivery workflow end-to-end. Hermes, messenger of the gods, carries the interface to the user; resolute because the design is minimal and unwavering.

---

SHA: 1dd5f40

v7.1.20260422 - Precise Hypatia

v7.1.20260422 - Precise Hypatia

"Precision of thought, clarity of exposition."

Changelog

Added

• docs: concrete sample content for all template documents — glossary, ADRs, discovery, scope journal, system overview (#107)
• docs: two ADRs (version-source, verbosity-validation) cross-referenced from system.md (#107)

Changed

• docs: system.md fully rewritten with Domain Model (Entities, Actions, Relationships), C4 Context + Container Mermaid diagrams, ADR Index, Completed Features (#107)
• docs: context.md and container.md deleted — C4 diagrams now live as sections inside system.md (SA-owned) (#107)
• docs: domain model section renamed "Actions" (was "Verbs") throughout (#107)
• chore: workflow and skill improvements — AGENTS.md, FLOW.md, architect, define-scope, implement, run-session, select-feature, update-docs, verify skills (#107)

Removed

• docs: standalone docs/context.md and docs/container.md (#107)
• template: domain-model.md.template removed from implement skill — domain model is a section of system.md (#107)

Summary

This release makes every template document concrete. All placeholder text has been replaced with a fully worked example grounded in the display-version feature. The system overview now integrates C4 Context and Container diagrams as sections rather than separate files. The name Precise Hypatia honours Hypatia of Alexandria — mathematician and philosopher — whose defining quality was the precision and clarity she brought to inherited knowledge.

---

SHA: bd67e19

v7.0.20260422 - Mighty Titan

v7.0.20260422 - Mighty Titan

"From a single template, a titan stirs — production Python, forged once, deployed everywhere."

Changelog

Added

• PyPI publishing pipeline: tag-release.yml auto-tags on version bump; pypi-publish.yml builds, publishes via OIDC, and creates GitHub Release (#100)
• FLOW.md + WORK.md: static state machine + dynamic work tracker replace TODO.md (#100)
• flow skill: full state machine protocol with auto-detection, templates, and recovery (#100)
• version-control skill: Git safety, branch lifecycle, --no-ff merge protocol (#100)
• architect skill: Step 2 architecture protocol, split from implement (#100)
• System Architect agent: owns Steps 2 and 4 — closed SA→SE→SA review loop (#100)
• Post-mortem protocol: compact template, fix/<stem> branch from original start commit (#100)
• Research library: entries 59–77 — arc42, Google design docs, Actor Model, Statecharts, Session Types, and more (#100)

Changed

• Project renamed to temple8: all user-authored files updated; git remote updated (#100)
• README rewritten: audience-first, painpoint-led, living documentation focus (#100)
• 8-glyph in logo and banner: ∞ in pediment replaced with 8 between the pillars (#100)
• SA/SE role split: reviewer absorbed into system-architect; SE owns Step 3 only (#100)
• pytest-beehave constraint: corrected to >=0.1,<1 matching published version (#101)

Removed

• TODO.md, .dockerignore, docs/architecture.md, reviewer agent (#100)

Summary

Mighty Titan marks temple8's identity — a major overhaul introducing the SA/SE closed review loop, the FLOW state machine, a PyPI publishing pipeline, and a visual rebrand with the 8-glyph between the temple pillars.

---

SHA: c3a8bd7

v6.4.20260420 - Minimal Prometheus

v6.4.20260420 - Minimal Prometheus

"Precision without ornamentation. Design without noise."

Changelog

Added

• Branding system: docs/branding.md — project identity, colour palette, release naming convention, and wording guidelines; agents read this file to personalise release names, C4 diagram colours, and docs without touching .opencode/ (#89)
• Designer agent (designer.md): owns docs/branding.md and docs/assets/; uses design-colors and design-assets skills (#89)
• design-colors skill: step-by-step colour palette selection with WCAG 2.1 AA 4.5:1 contrast validation; Itten/Albers colour theory embedded inline (#89)
• design-assets skill: SVG banner and logo creation workflow; W3C SVG 2 spec and WCAG 1.1.1 aria-label requirements embedded (#89)
• setup-project Step 6 Branding: collects tagline, mission, vision, tone, theme, and colours; suggests WCAG-validated palettes when user provides a theme but no colours; writes docs/branding.md (#89)
• Output Style + Rule #8 in run-session skill: minimalist output discipline — signal only, no tool narration, session ends with Next: line (#89)

Changed

• Skill renames to verb-noun convention: session-workflow → run-session, scope → define-scope, implementation → implement, feature-selection → select-feature, living-docs → update-docs, pr-management → create-pr, design-patterns → apply-patterns, code-quality → check-quality — all references updated across agents, skills, and AGENTS.md (#89)
• docs/images/ → docs/assets/: asset directory renamed; README.md path updated (#89)
• git-release v1.1: reads docs/branding.md for optional release naming and theme; release name omitted from commit/release title if convention is absent (#89)
• update-docs skill: reads docs/branding.md primary/accent colours to apply Mermaid C4 diagram theming (#89)

Summary

This release introduces a branding layer that lets each project instance personalise its agents' output — from release names to diagram colours — through a single docs/branding.md file, with no changes to .opencode/. A new designer agent and two design skills bring WCAG-validated colour and SVG asset workflows into the template. All 8 skills are renamed to verb-noun convention for consistency. The name Minimal Prometheus honours the Greek antiquity theme set in the template's own branding and the minimalist output discipline added to run-session.

---

SHA: 4cd2b8f

v6.3.20260419

Bump pytest-beehave to 3.3

v6.2.20260419 - Autonomous Stenella

v6.2.20260419 - Autonomous Stenella

"The workflow runs itself — IDs assigned, stubs generated, declarations enforced."

Changelog

Added

• pytest-beehave integration: @id tags now auto-assigned to untagged Example: blocks on every pytest run; test stubs auto-generated from .feature files at Step 2 end — no manual ID generation or stub writing required (#78)
• Self-declaration defense in depth: all 25 items numbered 1–25 in implementation/SKILL.md; verify/SKILL.md now hard-gates on completeness (count must equal 25, sequence must be gapless) before item audit begins (#78)

Changed

• Naming convention: .feature paths use <feature-stem> (kebab); test dirs use <feature_slug> (underscore) — applied across all skills, AGENTS.md, and docs (#78)
• conftest.py: removed manual deprecated skip hook — now owned by pytest-beehave (#78)
• scope/SKILL.md: removed all manual @id generation instructions and uniqueness checklist items (#78)
• product-owner.md: PO writes Example: blocks only — no @id authoring (#78)
• README: added "Why this template?" section; pytest-beehave in tooling table; auto-generation illustrated (#78)
• verify/SKILL.md report table: expanded Self-Declaration Audit from 21 to 25 numbered rows (#78)

Summary

This release makes the workflow autonomous at two key friction points: @id tag assignment and test stub generation now happen automatically when tests run, removing the most error-prone manual steps from the SE and PO roles. The numbered self-declaration with reviewer completeness enforcement closes the gap between a 21-item report table and the 25-item declaration template.

The name Autonomous Stenella (Stenella = spotted dolphin genus) reflects a workflow that navigates on its own.

---

SHA: dd9a6d6

v6.1.20260419 - Contextual Ambystoma

v6.1.20260419 - Contextual Ambystoma

"Living documentation, context-aware vocabulary, and a slimmer reviewer — all crystallised."

Changelog

Added

• feat: living-docs skill — C4 diagrams and glossary now updated as part of the release process (#77)

Fixed

• fix: correct stale Self-Declaration location references in verify skill (#77)

Changed

• refactor: slim reviewer.md and verify skill to reduce token load (#77)

Summary

This hotfix ships three post-release improvements to v6.1.20260419: the living-docs skill is now integrated into the git-release process (step 5), the verify skill's Self-Declaration audit section references are corrected, and the reviewer agent has been trimmed to reduce context overhead.

---

SHA: 633e701