diff --git a/.claude-plugin/marketplace.json b/.claude-plugin/marketplace.json index 1c85f24a..f2e0af8c 100644 --- a/.claude-plugin/marketplace.json +++ b/.claude-plugin/marketplace.json @@ -12,7 +12,7 @@ "name": "codingbuddy", "source": "./packages/claude-code-plugin", "description": "PLAN/ACT/EVAL workflow, specialist agents, and reusable skills for systematic TDD development", - "version": "5.1.3", + "version": "5.4.0", "category": "development" } ] diff --git a/CHANGELOG.md b/CHANGELOG.md index 008622c3..bc2ea51c 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -5,6 +5,31 @@ All notable changes to this project will be documented in this file. The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.1.0/), and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html). +## [5.4.0] - 2026-04-07 + +### Added + +#### Question-First Planning (Epic #1424) +- Clarification Gate — ambiguous prompts trigger one clarifying question before planning (#1423) +- Staged Planning — Discover→Design→Plan becomes the real default workflow (#1420) +- Planning Agent Guidance — question-first contract survives standard verbosity (#1419) +- Council Scene Rendering — real opening council scene in first response for PLAN/EVAL/AUTO (#1421) +- Permission Forecast — display upcoming permission classes before execution (#1418) +- Execution Gate — suppress dispatch-ready while planning is gated (#1422) + +#### Council State Pipeline (#1361, #1367, #1368) +- Seed request-level council state from UserPromptSubmit (#1361) +- Compact live council badges in PreToolUse statusMessage (#1367) +- Council handoff, stage, and blocker state from PostToolUse (#1368) + +#### EVAL Review Integration +- `review_pr` MCP tool for EVAL-mode PR review +- EVAL-mode review agent prompt template in adapter docs +- Integrated `review_pr` into parse_mode EVAL dispatch + +### Fixed +- Vite 7.3.1→8.0.5 security bump (GHSA-p9ff-h696-f583) + ## [5.2.0] - 2026-04-03 ### Added diff --git a/README.es.md b/README.es.md index 5cd878ce..69091dfc 100644 --- a/README.es.md +++ b/README.es.md @@ -250,17 +250,21 @@ npx codingbuddy mcp --tui --- -## Novedades en v5.1.2 +## Novedades en v5.4.0 -**StatusLine HUD** — Métricas de sesión en tiempo real en la UI de Claude Code: modo, costo, tasa de caché, uso de contexto. +**Planificación con preguntas primero** — Codingbuddy ahora pregunta antes de planificar. Los prompts ambiguos activan una pregunta de clarificación, y los prompts claros avanzan por las etapas Discover→Design→Plan con confirmación del usuario en cada paso. -``` -◕‿◕ CB v5.1.2 | PLAN 🟢 | 12m | ~$0.23 | Cache:87% | Ctx:45% -``` +**Escena del consejo** — Los modos PLAN, EVAL y AUTO comienzan con una escena de apertura que muestra qué agentes especialistas están ensamblados y listos para colaborar. + +**Pronóstico de permisos** — Antes de la ejecución, se muestran las clases de permisos necesarias (repo-write, network, external) para preparar los paquetes de aprobación con anticipación. -**Barra lateral tmux** — Ejecute Claude Code dentro de tmux y obtenga un panel TUI automático como barra lateral. +**Puerta de ejecución** — Cuando la etapa de planificación no ha avanzado más allá del descubrimiento, se suprime el despacho de especialistas para evitar trabajo prematuro. -**Validación de lanzamiento** — Nuevas herramientas MCP: `validate_plugin_manifest` y `pre_release_check`. +**Pipeline de estado del consejo** — Insignias en tiempo real muestran transferencias de agentes, transiciones de etapa y bloqueadores durante la ejecución de herramientas. + +``` +◕‿◕ CB v5.4.0 | PLAN 🟢 | 12m | ~$0.23 | Cache:87% | Ctx:45% +``` --- diff --git a/README.ja.md b/README.ja.md index 81a4ddc0..381bf833 100644 --- a/README.ja.md +++ b/README.ja.md @@ -250,17 +250,21 @@ npx codingbuddy mcp --tui --- -## v5.1.2の新機能 +## v5.4.0の新機能 -**StatusLine HUD** — Claude Code UIにリアルタイムセッション指標を表示:モード、コスト、キャッシュ率、コンテキスト使用量。 +**質問優先プランニング** — Codingbuddyが計画前に質問します。曖昧なプロンプトは明確化の質問をトリガーし、明確なプロンプトはDiscover→Design→Planの段階をユーザー確認とともに進行します。 -``` -◕‿◕ CB v5.1.2 | PLAN 🟢 | 12m | ~$0.23 | Cache:87% | Ctx:45% -``` +**カウンシルシーン** — PLAN、EVAL、AUTOモードで、どの専門エージェントが構成されているかを示すオープニングシーンから始まります。 + +**権限予測** — 実行前に必要な権限クラス(repo-write、network、external)を事前に表示します。 -**tmuxサイドバー** — tmux内でClaude Codeを実行すると、TUIダッシュボードサイドバーが自動設定されます。 +**実行ゲート** — プランニング段階が探索以前の場合、専門家のディスパッチを抑制して早期作業を防ぎます。 -**リリース検証** — `validate_plugin_manifest`と`pre_release_check` MCPツールを追加。 +**カウンシル状態パイプライン** — ツール実行中にエージェントハンドオフ、段階遷移、ブロッカーをリアルタイムバッジで表示します。 + +``` +◕‿◕ CB v5.4.0 | PLAN 🟢 | 12m | ~$0.23 | Cache:87% | Ctx:45% +``` --- diff --git a/README.ko.md b/README.ko.md index bc33580c..61b6a400 100644 --- a/README.ko.md +++ b/README.ko.md @@ -250,17 +250,21 @@ npx codingbuddy mcp --tui --- -## v5.1.2 새로운 기능 +## v5.4.0 새로운 기능 -**StatusLine HUD** — Claude Code UI에 실시간 세션 지표 표시: 현재 모드, 예상 비용, 캐시 적중률, 컨텍스트 사용량. +**질문-우선 플래닝** — Codingbuddy가 이제 계획 전에 먼저 질문합니다. 모호한 프롬프트는 명확화 질문을 트리거하고, 명확한 프롬프트는 Discover→Design→Plan 단계를 사용자 확인과 함께 진행합니다. -``` -◕‿◕ CB v5.1.2 | PLAN 🟢 | 12m | ~$0.23 | Cache:87% | Ctx:45% -``` +**카운슬 씬** — PLAN, EVAL, AUTO 모드에서 어떤 전문 에이전트가 구성되었는지 보여주는 오프닝 씬으로 시작합니다. + +**권한 예측** — 실행 전에 필요한 권한 클래스(repo-write, network, external)를 미리 표시하여 승인 준비를 할 수 있습니다. -**tmux 사이드바** — tmux 안에서 Claude Code를 실행하면 TUI 대시보드 사이드바가 자동으로 설정됩니다. +**실행 게이트** — 플래닝 단계가 탐색 이전이면 전문가 디스패치를 억제하여 조기 작업을 방지합니다. -**릴리즈 검증** — `validate_plugin_manifest`(plugin.json 스키마 검증)와 `pre_release_check`(에코시스템 자동 감지 릴리즈 사전 검증) MCP 도구 추가. +**카운슬 상태 파이프라인** — 도구 실행 중 에이전트 핸드오프, 단계 전환, 블로커를 실시간 뱃지로 표시합니다. + +``` +◕‿◕ CB v5.4.0 | PLAN 🟢 | 12m | ~$0.23 | Cache:87% | Ctx:45% +``` --- diff --git a/README.md b/README.md index 699d8805..2abdc147 100644 --- a/README.md +++ b/README.md @@ -100,21 +100,17 @@ Start coding with `PLAN`, `ACT`, `EVAL`, or `AUTO` keywords. --- -## What's New in v5.2.0 +## What's New in v5.4.0 -**PR Guardian** — `pr_quality_report` MCP tool runs specialist agents on changed files, integrated with `/ship` quality gate for automated PR quality checks. +**Question-First Planning** — Codingbuddy now asks before it plans. Ambiguous prompts trigger a clarifying question, and clear prompts walk through Discover→Design→Plan stages with explicit user confirmation at each step. -**Context Rescue** — `create_briefing` and `resume_session` tools enable cross-session recovery, preserving decisions and context across conversation boundaries. +**Council Scene** — PLAN, EVAL, and AUTO modes open with a visible council scene showing which specialist agents are assembled and ready to collaborate. -**Session Intelligence** — SessionAnalyzer generates Intelligence Reports in the stop hook, providing actionable insights from every coding session. +**Permission Forecast** — Before execution, codingbuddy surfaces what permission classes (repo-write, network, external) will be needed, so you can prepare approval bundles in advance. -**Rule Impact Dashboard** — Event tracking, stats writer, and impact reporting measure which rules actually prevent issues — proving your AI coding is improving. +**Execution Gate** — When the planning stage hasn't advanced past discovery, specialist dispatch is suppressed to avoid premature work. -**Plugin Marketplace** — Full plugin lifecycle with `install`, `search`, `update`, `plugins`, and `uninstall` CLI commands, plus registry resolver for discovering community plugins. - -**Skill Discovery** — Filesystem-based `list_skills` with `triggers` frontmatter in SKILL.md for automatic skill activation based on context. - -**TUI** — Single multi-session model with manual `codingbuddy tui` launch. Auto-open removed for cleaner MCP server startup. +**Council State Pipeline** — Real-time council badges show agent handoffs, stage transitions, and blockers as they happen during tool execution. --- diff --git a/README.zh-CN.md b/README.zh-CN.md index 207c74ad..3d952d05 100644 --- a/README.zh-CN.md +++ b/README.zh-CN.md @@ -250,17 +250,21 @@ npx codingbuddy mcp --tui --- -## v5.1.2 新功能 +## v5.4.0 新功能 -**StatusLine HUD** — 在Claude Code UI中实时显示会话指标:模式、成本、缓存率、上下文使用量。 +**问题优先规划** — Codingbuddy 现在在规划之前先提问。模糊的提示会触发澄清问题,清晰的提示会通过 Discover→Design→Plan 阶段逐步推进,每步都需用户确认。 -``` -◕‿◕ CB v5.1.2 | PLAN 🟢 | 12m | ~$0.23 | Cache:87% | Ctx:45% -``` +**议会场景** — PLAN、EVAL 和 AUTO 模式以展示哪些专家代理已组建的开场场景开始。 + +**权限预测** — 执行前显示即将需要的权限类别(repo-write、network、external),便于提前准备审批。 -**tmux侧边栏** — 在tmux中运行Claude Code时,自动配置TUI仪表板侧边栏。 +**执行门控** — 当规划阶段尚未进入设计阶段时,抑制专家调度以避免过早执行。 -**发布验证** — 新增`validate_plugin_manifest`和`pre_release_check` MCP工具。 +**议会状态管道** — 在工具执行期间,实时徽章显示代理交接、阶段转换和阻塞器。 + +``` +◕‿◕ CB v5.4.0 | PLAN 🟢 | 12m | ~$0.23 | Cache:87% | Ctx:45% +``` --- diff --git a/apps/mcp-server/package.json b/apps/mcp-server/package.json index e71c6af9..3c8d8fc5 100644 --- a/apps/mcp-server/package.json +++ b/apps/mcp-server/package.json @@ -1,6 +1,6 @@ { "name": "codingbuddy", - "version": "5.3.0", + "version": "5.4.0", "description": "Multi-AI Rules MCP Server - One source of truth for AI coding rules across all AI assistants", "author": "JeremyDev87", "license": "MIT", diff --git a/apps/mcp-server/src/shared/version.ts b/apps/mcp-server/src/shared/version.ts index cc9965a2..31bef326 100644 --- a/apps/mcp-server/src/shared/version.ts +++ b/apps/mcp-server/src/shared/version.ts @@ -2,4 +2,4 @@ * Single source of truth for the runtime package version. * Updated automatically by scripts/bump-version.sh on each release. */ -export const VERSION = '5.3.0'; +export const VERSION = '5.4.0'; diff --git a/packages/claude-code-plugin/.claude-plugin/plugin.json b/packages/claude-code-plugin/.claude-plugin/plugin.json index ddcb2e9d..31610cf5 100644 --- a/packages/claude-code-plugin/.claude-plugin/plugin.json +++ b/packages/claude-code-plugin/.claude-plugin/plugin.json @@ -1,6 +1,6 @@ { "name": "codingbuddy", - "version": "5.3.0", + "version": "5.4.0", "description": "PLAN/ACT/EVAL workflow with auto-detection, specialist agents, and reusable skills for systematic TDD development", "author": { "name": "JeremyDev87", diff --git a/packages/claude-code-plugin/README.md b/packages/claude-code-plugin/README.md index 7101b82b..84911e97 100644 --- a/packages/claude-code-plugin/README.md +++ b/packages/claude-code-plugin/README.md @@ -2,7 +2,7 @@ # CodingBuddy Claude Code Plugin -> Version 5.3.0 +> Version 5.4.0 Multi-AI Rules for consistent coding practices - PLAN/ACT/EVAL workflow, specialist agents, and reusable skills for systematic development. diff --git a/packages/claude-code-plugin/docs/bootstrap-architecture.md b/packages/claude-code-plugin/docs/bootstrap-architecture.md index fca21ecd..f9f15ff8 100644 --- a/packages/claude-code-plugin/docs/bootstrap-architecture.md +++ b/packages/claude-code-plugin/docs/bootstrap-architecture.md @@ -1,6 +1,6 @@ # Plugin Hook Bootstrap Architecture -> **Status:** current as of `codingbuddy-claude-plugin` v5.3.0 +> **Status:** current as of `codingbuddy-claude-plugin` v5.4.0 > **Related issues:** [#1380](https://github.com/JeremyDev87/codingbuddy/issues/1380) (this document), [#1376](https://github.com/JeremyDev87/codingbuddy/issues/1376) (parent), [#1381](https://github.com/JeremyDev87/codingbuddy/issues/1381) (stale legacy hook migration) This document explains **how** and **where** the CodingBuddy Claude Code plugin registers its hooks — and in particular, why `UserPromptSubmit` is nowhere to be found in `hooks/hooks.json`. diff --git a/packages/claude-code-plugin/docs/council-surface-model.md b/packages/claude-code-plugin/docs/council-surface-model.md index c2a4e526..1ce47045 100644 --- a/packages/claude-code-plugin/docs/council-surface-model.md +++ b/packages/claude-code-plugin/docs/council-surface-model.md @@ -146,7 +146,7 @@ Refreshed continuously by Claude Code. ``` Line 1 (always): -◕‿◕ CB v5.3.0 | PLAN 🟢 | 12m | ~$0.42 | ♻800/1.5k | Ctx:45% +◕‿◕ CB v5.4.0 | PLAN 🟢 | 12m | ~$0.42 | ♻800/1.5k | Ctx:45% Line 2 (conditional — only when agent is active): [◮ secu] [auth flow] [✓] @@ -157,7 +157,7 @@ Line 2 (conditional — only when agent is active): | Segment | Source | Example | |---------|--------|---------| | Buddy face | Constant | `◕‿◕` | -| Version | `installed_plugins.json` / HUD state | `CB v5.3.0` | +| Version | `installed_plugins.json` / HUD state | `CB v5.4.0` | | Mode | `hud-state.currentMode` | `PLAN` | | Health | Context usage thresholds | `🟢` (<60%), `🟡` (60-85%), `🔴` (>85%) | | Duration | stdin `cost.total_duration_ms` / HUD state | `12m` | diff --git a/packages/claude-code-plugin/docs/status-bar-model.md b/packages/claude-code-plugin/docs/status-bar-model.md index b6c77f4d..9ffaba02 100644 --- a/packages/claude-code-plugin/docs/status-bar-model.md +++ b/packages/claude-code-plugin/docs/status-bar-model.md @@ -20,13 +20,13 @@ Displays session metrics computed from stdin data and HUD state. **Format:** ``` -◕‿◕ CB v5.3.0 | PLAN 🟢 | 12m | ~$0.42 | ♻800/1.5k | Ctx:45% +◕‿◕ CB v5.4.0 | PLAN 🟢 | 12m | ~$0.42 | ♻800/1.5k | Ctx:45% ``` | Segment | Source | Description | | --------------- | ------------------------------- | ------------------------------------------ | | `◕‿◕` | Constant | Buddy face | -| `CB v5.3.0` | `hud_state.version` | Plugin version | +| `CB v5.4.0` | `hud_state.version` | Plugin version | | `PLAN` | `hud_state.currentMode` | Current workflow mode (PLAN/ACT/EVAL/AUTO) | | `🟢` | Computed from `ctx_pct` | Health indicator (see below) | | `12m` | `hud_state.sessionStartTimestamp`| Session duration | @@ -190,34 +190,34 @@ If any exception occurs during rendering, the script outputs a minimal fallback: ### PLAN Mode (No Agent) ``` -◕‿◕ CB v5.3.0 | PLAN 🟢 | 5m | ~$0.12 | ♻1.2k/3k | Ctx:22% +◕‿◕ CB v5.4.0 | PLAN 🟢 | 5m | ~$0.12 | ♻1.2k/3k | Ctx:22% ``` ### ACT Mode (With Agent) ``` -◕‿◕ CB v5.3.0 | ACT 🟢 | 18m | ~$1.05 | ♻8k/13k | Ctx:48% +◕‿◕ CB v5.4.0 | ACT 🟢 | 18m | ~$1.05 | ♻8k/13k | Ctx:48% 🤖 frontend-developer ``` ### EVAL Mode (High Context) ``` -◕‿◕ CB v5.3.0 | EVAL 🟡 | 45m | ~$3.20 | ♻24k/34k | Ctx:73% +◕‿◕ CB v5.4.0 | EVAL 🟡 | 45m | ~$3.20 | ♻24k/34k | Ctx:73% 🤖 security-specialist ``` ### AUTO Mode (Critical Context) ``` -◕‿◕ CB v5.3.0 | AUTO 🔴 | 1h12m | ~$8.50 | ♻95k/172k | Ctx:91% +◕‿◕ CB v5.4.0 | AUTO 🔴 | 1h12m | ~$8.50 | ♻95k/172k | Ctx:91% 🤖 code-quality-specialist ``` ### No Mode Set (Initial State — Cache Segment Hidden) ``` -◕‿◕ CB v5.3.0 | Ready 🟢 | 0m | ~$0.00 | Ctx:0% +◕‿◕ CB v5.4.0 | Ready 🟢 | 0m | ~$0.00 | Ctx:0% ``` > The cache segment is **omitted** in the initial state because no API calls have been made yet — there is no `current_usage` data to render. This is the documented fallback behavior, not a bug. diff --git a/packages/claude-code-plugin/package.json b/packages/claude-code-plugin/package.json index 6c35f4b2..bb792b0b 100644 --- a/packages/claude-code-plugin/package.json +++ b/packages/claude-code-plugin/package.json @@ -1,6 +1,6 @@ { "name": "codingbuddy-claude-plugin", - "version": "5.3.0", + "version": "5.4.0", "description": "Claude Code Plugin for CodingBuddy - PLAN/ACT/EVAL workflow, specialist agents, and reusable skills", "author": "JeremyDev87", "license": "MIT", @@ -53,7 +53,7 @@ "test:coverage": "vitest run --coverage" }, "peerDependencies": { - "codingbuddy": "^5.3.0" + "codingbuddy": "^5.4.0" }, "peerDependenciesMeta": { "codingbuddy": { diff --git a/packages/rules/package.json b/packages/rules/package.json index 1bc50752..55922977 100644 --- a/packages/rules/package.json +++ b/packages/rules/package.json @@ -1,6 +1,6 @@ { "name": "codingbuddy-rules", - "version": "5.3.0", + "version": "5.4.0", "description": "AI coding rules for consistent practices across AI assistants", "main": "index.js", "types": "index.d.ts", diff --git a/yarn.lock b/yarn.lock index cf24ec1d..88a596f0 100644 --- a/yarn.lock +++ b/yarn.lock @@ -5654,7 +5654,7 @@ __metadata: typescript: "npm:5.9.3" vitest: "npm:4.0.17" peerDependencies: - codingbuddy: ^5.3.0 + codingbuddy: ^5.4.0 peerDependenciesMeta: codingbuddy: optional: true