diff --git a/CHANGELOG.md b/CHANGELOG.md new file mode 100644 index 00000000..e0ded097 --- /dev/null +++ b/CHANGELOG.md @@ -0,0 +1,45 @@ +# Changelog + +## 3.0.0-SNAPSHOT — In development + +3.0.0-SNAPSHOT upgrades to Java 21, Spring Boot 4 and Spring AI 2. AgentScope 2 drives ReAct and plan-step execution loops; MateClaw retains approvals, permissions, Goals, SSE and persistence. This is a development snapshot. + +- MyBatis Plus 3.5.17 with the Boot 4 starter; MCP SDK 0.18.3. +- Native DashScope chat/embedding adapters; native speech SDK retained. +- Owned HTTP cancellation, guarded tool batches and durable Goal checkpoints prevent blind side-effect replay. +- Existing frontend interaction contracts and conversation history remain supported. + +[English details](mateclaw-server/src/main/resources/docs/en/changelog.md) · [中文说明](mateclaw-server/src/main/resources/docs/zh/changelog.md) + +## Cumulative changes since 2.3.0 + +3.0.0-SNAPSHOT includes the former 2.4.0 development cycle; 2.4.0 was not published as a separate stable release. The groups below cover capabilities added before the architecture migration as well. + +### Workspaces and conversations + +- Optional member file isolation resolves personal files by member identity with shared-file fallback; existing workspace roles and default sharing still apply (#644). +- Persistent channel conversations expire stale query context. Uploaded images resolve from conversation storage; directory uploads retain directory metadata and validate containment. +- Employee drafts accept JSON following reasoning blocks, and explicit tool opt-out is respected even for general capabilities (#648, #655). + +### Structured decisions, Goals, plans and teams + +- A bounded typed Choice / Boolean / Score decision pipeline adds durable audit, OFF / SHADOW / ACTIVE modes and scenario overrides. SHADOW is the default; suggestions cannot bypass domain constraints. +- Goal continuation, plan employee routing and team worker-result judgment share the decision foundation. Domain transactions commit outcomes and state; delayed decisions validate identity and fences. +- Richer routing context, worker contracts, delegated-step result checks and bounded Goal recovery retain existing completion gates and record applied follow-up suppression. + +### Models and DeepSeek Harness + +- DSH moves to the SDK runtime with pinned versions and controlled upgrades. Failed health checks invalidate cached health and broken settings remain editable (#650). +- Model probes validate completed responses rather than accepting connection success or empty output (#658). + +### Read-aloud, chat and tool presentation + +- Edge and DashScope read-aloud are restored, with provider voice lists, selection and preview (#646). +- Structured MCP results get dedicated presentation below the final answer. Ordinary tool results no longer automatically become final-answer text (#651, #656, #661). + +### Channels and database operations + +- Application-owned Feishu WebSocket readiness and lifecycle release owned resources on shutdown and reconnect (#659). +- Unsupported webhook channels are rejected before configuration persistence (#654). +- Flyway validation failures block startup by default; history repair requires explicit opt-in, with backup and recovery guidance (#657). +- Commercial-edition overview and links are added; commercial modules are not included in the open-source edition. diff --git a/README.md b/README.md index 690688e7..6a12d115 100644 --- a/README.md +++ b/README.md @@ -1,5 +1,10 @@
+> Current development version: **3.0.0-SNAPSHOT**. Java 21 / Spring Boot 4.1.1 / Spring AI 2.0.1 / AgentScope 2.0.4. AgentScope drives execution loops; MateClaw retains tool approvals, permissions, Goals, SSE and persistence. See Releases for stable builds. +> [3.0.0 architecture changelog](CHANGELOG.md) +> Includes all changes since 2.3.0: structured decisions, member file isolation, DSH SDK upgrades, voice preview and structured MCP presentation. + +

MateClaw Logo

diff --git a/README_zh.md b/README_zh.md index 6473e6c8..cdbe6a42 100644 --- a/README_zh.md +++ b/README_zh.md @@ -1,5 +1,10 @@
+> 当前开发版本：**3.0.0-SNAPSHOT**。Java 21 / Spring Boot 4.1.1 / Spring AI 2.0.1 / AgentScope 2.0.4。AgentScope 驱动执行循环，MateClaw 保留工具审批、权限、Goal、SSE 与持久化。正式发行版请查看 Releases。 +> [3.0.0 架构升级日志](CHANGELOG.md) +> 包含 2.3.0 之后全部累计更新：结构化决策、成员文件隔离、DSH SDK 升级、朗读音色试听和结构化 MCP 展示。 + +

MateClaw Logo

diff --git a/assets/architecture-biz-en.svg b/assets/architecture-biz-en.svg index d20d11a1..76330ea6 100644 --- a/assets/architecture-biz-en.svg +++ b/assets/architecture-biz-en.svg @@ -33,7 +33,7 @@ Digital Employee - Identity · Goal · Governance + Identity · Goal · Decision Native Runtime · DSH Runtime One policy + event plane diff --git a/assets/architecture-biz-zh.svg b/assets/architecture-biz-zh.svg index 1938471d..d1923871 100644 --- a/assets/architecture-biz-zh.svg +++ b/assets/architecture-biz-zh.svg @@ -34,7 +34,7 @@ 数字员工 - 身份 · 目标 · 治理 + 身份 · 目标 · 决策 Native Runtime · DSH Runtime 同一策略与事件平面 diff --git a/assets/architecture-tech-en.svg b/assets/architecture-tech-en.svg index c57f0088..c8b0aae5 100644 --- a/assets/architecture-tech-en.svg +++ b/assets/architecture-tech-en.svg @@ -81,20 +81,20 @@ Native Runtime - StateGraph · ReAct - Plan-Execute · Goals · Teams + AgentScope 2 · ReAct + Plan-step execution DSH Runtime Managed JSON-RPC Process - DeepSeek Harness · Cordis + SDK · Pinned upgrades Host Governance Workspace · Tool Guard - Approval · Credentials + Approval · Checkpoints @@ -110,7 +110,7 @@ - Spring Boot 3.5 + Spring Boot 4.1 App Framework Spring MVC + SSE @@ -158,7 +158,7 @@ - H2 / MySQL 8 + H2 · PG · MySQL Relational Storage @@ -194,7 +194,7 @@ - Java 21+ · Spring Boot 3.5 · Spring AI Alibaba · Vue 3 · TypeScript · Vite · Electron + 3.0.0-SNAPSHOT · Java 21 · Boot 4 · Spring AI 2 · AgentScope 2 · Vue 3 · Electron claw.mate.vip diff --git a/assets/architecture-tech-zh.svg b/assets/architecture-tech-zh.svg index 2ab7a073..2405dd5f 100644 --- a/assets/architecture-tech-zh.svg +++ b/assets/architecture-tech-zh.svg @@ -84,20 +84,20 @@ Native Runtime - StateGraph · ReAct - Plan-Execute · Goal · Team + AgentScope 2 · ReAct + 计划单步执行 DSH Runtime 受管理 JSON-RPC 进程 - DeepSeek Harness · Cordis + SDK · Pinned upgrades 宿主治理 Workspace · Tool Guard - 审批 · 凭证隔离 + 审批 · 检查点 · 决策 @@ -113,7 +113,7 @@ - Spring Boot 3.5 + Spring Boot 4.1 应用框架 Spring MVC + SSE @@ -161,7 +161,7 @@ - H2 / MySQL 8 + H2 · PG · MySQL 关系型存储 @@ -198,7 +198,7 @@ - Java 21+ · Spring Boot 3.5 · Spring AI Alibaba · Vue 3 · TypeScript · Vite · Electron + 3.0.0-SNAPSHOT · Java 21 · Boot 4 · Spring AI 2 · AgentScope 2 · Vue 3 · Electron claw.mate.vip diff --git a/mateclaw-desktop/electron/main/index.ts b/mateclaw-desktop/electron/main/index.ts index 5e288468..1d6c36c3 100644 --- a/mateclaw-desktop/electron/main/index.ts +++ b/mateclaw-desktop/electron/main/index.ts @@ -812,7 +812,7 @@ function showAboutDialog(): void { detail: [ `Version: ${app.getVersion()}`, '', - 'Your intelligent AI assistant powered by Spring AI Alibaba.', + 'Your intelligent AI assistant powered by Spring AI 2 + AgentScope 2.', '', `Copyright © 2026 MateClaw Team`, ].join('\n'), diff --git a/mateclaw-desktop/package.json b/mateclaw-desktop/package.json index f5c5a43f..9ae161e1 100644 --- a/mateclaw-desktop/package.json +++ b/mateclaw-desktop/package.json @@ -1,7 +1,7 @@ { "name": "mateclaw-desktop", - "version": "2.4.0-SNAPSHOT", - "description": "MateClaw Desktop - AI Assistant powered by Spring AI Alibaba", + "version": "3.0.0-SNAPSHOT", + "description": "MateClaw Desktop - AI Assistant powered by Spring AI 2 + AgentScope 2", "author": "MateClaw Team", "license": "Apache-2.0", "main": "dist-electron/main/index.js", diff --git a/mateclaw-server/src/main/resources/docs/en/agents.md b/mateclaw-server/src/main/resources/docs/en/agents.md index 5b6896d8..92a3da65 100644 --- a/mateclaw-server/src/main/resources/docs/en/agents.md +++ b/mateclaw-server/src/main/resources/docs/en/agents.md @@ -508,3 +508,7 @@ You'll see node-by-node execution: state transitions, dispatcher routing, iterat - [Workflow](./workflow) (1.3.0+) — orchestrate multiple digital employees and system actions into a business process - [Triggers](./triggers) (1.3.0+) — let events automatically start workflows or agent conversations - [Architecture](./architecture) — native ReAct and plan runtime in depth + +## Execution and tool boundaries (3.0.0) + +AgentScope 2 drives native ReAct and plan-step loops. MateClaw retains actual tool execution through approvals, permissions and audit. Explicit tool opt-out remains effective and employee-draft JSON may follow reasoning blocks. diff --git a/mateclaw-server/src/main/resources/docs/en/changelog.md b/mateclaw-server/src/main/resources/docs/en/changelog.md new file mode 100644 index 00000000..3986daad --- /dev/null +++ b/mateclaw-server/src/main/resources/docs/en/changelog.md @@ -0,0 +1,57 @@ +# 3.0.0-SNAPSHOT: execution architecture upgrade + +Development snapshot; no stable 3.0.0 release has been published. + +## Cumulative changes since 2.3.0 + +3.0.0-SNAPSHOT includes the former 2.4.0 development cycle; 2.4.0 was not published as a separate stable release. The groups below cover capabilities added before the architecture migration as well. + +### Workspaces and conversations + +- Optional member file isolation resolves personal files by member identity with shared-file fallback; existing workspace roles and default sharing still apply (#644). +- Persistent channel conversations expire stale query context. Uploaded images resolve from conversation storage; directory uploads retain directory metadata and validate containment. +- Employee drafts accept JSON following reasoning blocks, and explicit tool opt-out is respected even for general capabilities (#648, #655). + +### Structured decisions, Goals, plans and teams + +- A bounded typed Choice / Boolean / Score decision pipeline adds durable audit, OFF / SHADOW / ACTIVE modes and scenario overrides. SHADOW is the default; suggestions cannot bypass domain constraints. +- Goal continuation, plan employee routing and team worker-result judgment share the decision foundation. Domain transactions commit outcomes and state; delayed decisions validate identity and fences. +- Richer routing context, worker contracts, delegated-step result checks and bounded Goal recovery retain existing completion gates and record applied follow-up suppression. + +### Models and DeepSeek Harness + +- DSH moves to the SDK runtime with pinned versions and controlled upgrades. Failed health checks invalidate cached health and broken settings remain editable (#650). +- Model probes validate completed responses rather than accepting connection success or empty output (#658). + +### Read-aloud, chat and tool presentation + +- Edge and DashScope read-aloud are restored, with provider voice lists, selection and preview (#646). +- Structured MCP results get dedicated presentation below the final answer. Ordinary tool results no longer automatically become final-answer text (#651, #656, #661). + +### Channels and database operations + +- Application-owned Feishu WebSocket readiness and lifecycle release owned resources on shutdown and reconnect (#659). +- Unsupported webhook channels are rejected before configuration persistence (#654). +- Flyway validation failures block startup by default; history repair requires explicit opt-in, with backup and recovery guidance (#657). +- Commercial-edition overview and links are added; commercial modules are not included in the open-source edition. + +## Platform and execution core + +Java 21, Spring Boot 4.1.1, Spring AI 2.0.1 and AgentScope Java 2.0.4. Spring AI Alibaba starter and Graph dependencies are removed. AgentScope drives ordinary ReAct and the reasoning/tool/observation loop inside each Plan-Execute step. MateClaw retains overall plan scheduling, Goal budgets and acceptance, and team coordination. + +ApplicationGuardedToolkit delegates tool batches to the application executor, retaining approvals, workspace permissions, audit and cancellation. Model requests use existing Spring AI/provider adapters; MateClaw owns SSE, history and persistence. + +## Models, tools and recovery + +- Native DashScope chat and embedding adapters replace Alibaba AI integration; the native speech SDK remains. +- OpenAI/Anthropic adapters preserve reasoning, media, cache and usage. Cancellation aborts owned HTTP calls before reader cleanup and prevents subsequent failover. +- MCP SDK 0.18.3 preserves call IDs, structured results and protocol errors; local execution capabilities stay out of wire metadata. +- Goal tool checkpoints persist execution identity before side effects. Uncertain outcomes block blind replay, retaining approval recovery and session isolation. +- MyBatis Plus 3.5.17 uses the Spring Boot 4 starter; string IDs and existing conversation contracts are retained. +- Binary skill bundles and bundled documentation synchronization are hardened. + +## Upgrade and validation + +Stop the old process and back up the database, workspace, plugins and configuration before upgrading. Build with Java 21 and check third-party plugin compatibility. Rollback uses the pre-upgrade backup; downgrading an upgraded database is not verified. + +Validation covers the Maven reactor, frontend build, PostgreSQL/MySQL migrations and transactional pagination, approvals/cancellation, process-crash recovery and six-hour sustained execution. A real Kingbase instance was not tested. Existing frontend interaction contracts remain compatible; this migration does not rewrite the chat UI. diff --git a/mateclaw-server/src/main/resources/docs/en/channels.md b/mateclaw-server/src/main/resources/docs/en/channels.md index de485d82..0d3b961a 100644 --- a/mateclaw-server/src/main/resources/docs/en/channels.md +++ b/mateclaw-server/src/main/resources/docs/en/channels.md @@ -716,3 +716,7 @@ Cron edits now retain both delivery channel and target, so changing a schedule o - [Chat & Messaging](./chat) — message flow, segments, streaming events - [Agents](./agents) — what's actually answering through each channel - [Configuration](./config) — global channel tuning + +## Channel configuration and lifecycle (3.0.0) + +The application owns Feishu WebSocket readiness, reconnect and shutdown. Unsupported webhook channel types are rejected before saving; stale query context is cleared from persistent conversations to avoid affecting later tasks. diff --git a/mateclaw-server/src/main/resources/docs/en/chat.md b/mateclaw-server/src/main/resources/docs/en/chat.md index f76e783a..ed78d8b2 100644 --- a/mateclaw-server/src/main/resources/docs/en/chat.md +++ b/mateclaw-server/src/main/resources/docs/en/chat.md @@ -416,3 +416,7 @@ Sessions can batch-delete up to 200 selected conversations. The server deduplica - [Memory](./memory) — what persists across conversations - [Channels](./channels) — every other place chat can happen - [LLM Wiki](./wiki) — what the agent can read while answering + +## Structured tool results (3.0.0) + +Structured MCP results are displayed separately below the final answer. Ordinary tool output is not automatically appended to answer text; call IDs and structured content remain in streaming events and history. Images resolve from conversation storage and directory attachments retain directory information. diff --git a/mateclaw-server/src/main/resources/docs/en/goals.md b/mateclaw-server/src/main/resources/docs/en/goals.md index 8b25bf24..bf16256b 100644 --- a/mateclaw-server/src/main/resources/docs/en/goals.md +++ b/mateclaw-server/src/main/resources/docs/en/goals.md @@ -449,3 +449,7 @@ Two tables, all `mate_`-prefixed: **A goal isn't a new feature on the worker. It's a state change.** Before, the worker forgot the moment it answered. Goals make a worker remember one thing across many turns — what it's working on, what's still missing, when it counts as done. You say it once. The ring next to the avatar tracks the rest. + +## 3.0.0: continuation decisions and side-effect recovery + +Continuation suggestions use bounded structured decisions and domain constraints, with SHADOW as default. Fenced checkpoints are saved before tool execution; uncertain outcomes after side effects require intervention rather than blind replay. See [decision layer](./decision-layer) and [changelog](./changelog). diff --git a/mateclaw-server/src/main/resources/docs/en/index.md b/mateclaw-server/src/main/resources/docs/en/index.md index 632ab655..eb517802 100644 --- a/mateclaw-server/src/main/resources/docs/en/index.md +++ b/mateclaw-server/src/main/resources/docs/en/index.md @@ -19,10 +19,13 @@ hero: text: GitHub link: https://github.com/mateaix/mateclaw - theme: alt - text: v2.3.0 Release Notes - link: /en/releases/2.3.0 + text: 3.0.0 Snapshot Changelog + link: /en/releases/3.0.0 features: + - icon: ⚙️ + title: 3.0.0-SNAPSHOT — Execution architecture upgrade + details: 3.0.0-SNAPSHOT upgrades to Java 21, Spring Boot 4 and Spring AI 2. AgentScope 2 drives ReAct and plan-step execution loops; MateClaw retains approvals, permissions, Goals, SSE and persistence. This is a development snapshot. - icon: ✅ title: 2.3.0 — Managed JSON acceptance details: Users select required top-level fields; the server binds checks to current requirements and immutable JSON versions. Execution evidence defaults to observation, and ordinary artifact diagnostics do not replace acceptance bindings. diff --git a/mateclaw-server/src/main/resources/docs/en/intro.md b/mateclaw-server/src/main/resources/docs/en/intro.md index d025fc9a..ad7521c7 100644 --- a/mateclaw-server/src/main/resources/docs/en/intro.md +++ b/mateclaw-server/src/main/resources/docs/en/intro.md @@ -7,6 +7,11 @@ head: content: MateClaw,multi-agent AI,self-hosted AI,AI operating system,Spring AI + AgentScope,ReAct,Plan-and-Execute,MCP,LLM Wiki,memory lifecycle,Tool Guard,open source --- +3.0.0-SNAPSHOT upgrades to Java 21, Spring Boot 4 and Spring AI 2. AgentScope 2 drives ReAct and plan-step execution loops; MateClaw retains approvals, permissions, Goals, SSE and persistence. This is a development snapshot. + +[Upgrade guidance and changelog](./changelog) + + # MateClaw — Self-hosted Multi-Agent AI Operating System **v2.3.0 (released 2026-09-20) — JSON acceptance and execution evidence for persistent Goals.** Users can configure required JSON fields and bind checks to current requirements and immutable artifact versions. This release also adds controlled worker intervention and skill document/folder uploads, with fixes for DSH history and output budgets, vLLM reasoning, and SSE framing. [v2.3.0](./releases/2.3.0) diff --git a/mateclaw-server/src/main/resources/docs/en/mcp.md b/mateclaw-server/src/main/resources/docs/en/mcp.md index 2fe0bce5..831a1797 100644 --- a/mateclaw-server/src/main/resources/docs/en/mcp.md +++ b/mateclaw-server/src/main/resources/docs/en/mcp.md @@ -647,3 +647,7 @@ registerToolResultRenderer('inventory', { ``` The component receives a `data` prop. A tool can then return `{ "type": "inventory", "data": { "name": "Example", "quantity": 10 } }` inside `blocks`. Registration requires a frontend rebuild; tool responses can only name registered types and cannot import components or execute scripts. Custom components must validate their schema and safely render untrusted values. + +## 3.0.0: protocol and presentation boundaries + +SDK 0.18.3 preserves call IDs, structured content and protocol errors. Ordinary tool results are presented separately below final answers. Transport failures with uncertain side effects do not trigger automatic duplicate execution. diff --git a/mateclaw-server/src/main/resources/docs/en/models.md b/mateclaw-server/src/main/resources/docs/en/models.md index 414bcabd..450a40ae 100644 --- a/mateclaw-server/src/main/resources/docs/en/models.md +++ b/mateclaw-server/src/main/resources/docs/en/models.md @@ -526,3 +526,7 @@ The OpenAI-compatible path also preserves `integer` / `number` in tool JSON Sche - [Configuration](./config) — full config reference - [Agents](./agents) — how agents use models - [Admin Console](./console) — UI for model management + +## Model probes and cancellation (3.0.0) + +A probe requires a valid completed response; connection success or empty output does not prove model availability. Stopping generation cancels HTTP calls owned by the request and prevents subsequent failover. diff --git a/mateclaw-server/src/main/resources/docs/en/multimodal.md b/mateclaw-server/src/main/resources/docs/en/multimodal.md index ca231aec..18e9aba2 100644 --- a/mateclaw-server/src/main/resources/docs/en/multimodal.md +++ b/mateclaw-server/src/main/resources/docs/en/multimodal.md @@ -197,3 +197,7 @@ In short: configure a vision model under **Settings → Models → Multimodal si - [Chat & Messaging](./chat) — attachment input, multimodal sidecar routing, how generated media attaches to messages - [Models](./models) — provider configuration UI, multimodal sidecar settings - [Tools](./tools) — the tool system that hosts multimodal generation + +## Read-aloud voices and preview (3.0.0) + +Chat read-aloud supports Edge and DashScope. Provider voice lists support selection and preview; availability depends on provider configuration. APIs: GET /api/v1/tts/voices, POST /api/v1/tts/preview and POST /api/v1/tts/synthesize. diff --git a/mateclaw-server/src/main/resources/docs/en/quickstart.md b/mateclaw-server/src/main/resources/docs/en/quickstart.md index d81e0153..ddeebac3 100644 --- a/mateclaw-server/src/main/resources/docs/en/quickstart.md +++ b/mateclaw-server/src/main/resources/docs/en/quickstart.md @@ -1,5 +1,8 @@ # Quick Start +The source development branch is `3.0.0-SNAPSHOT`; stable installers remain the versions published in Releases. Read the [upgrade guidance](./changelog#upgrade-and-validation) before migrating. + + Sixty seconds to first message. **One path. Desktop app.** If you want Docker or local development instead, those live in [Configuration](./config) and [Contributing](./contributing). This page does one thing only — get you from zero to a working agent as fast as humanly possible. diff --git a/mateclaw-server/src/main/resources/docs/en/releases.md b/mateclaw-server/src/main/resources/docs/en/releases.md index 701168b3..d78b269c 100644 --- a/mateclaw-server/src/main/resources/docs/en/releases.md +++ b/mateclaw-server/src/main/resources/docs/en/releases.md @@ -10,6 +10,7 @@ For historical diffs, check the corresponding git tag. For the "why" behind a fe | Version | Date | Highlights | |---------|------|------------| +| [3.0.0-SNAPSHOT](./releases/3.0.0) | In development | Java 21 · Boot 4 · Spring AI 2 · AgentScope 2 execution · structured decisions · member files · DSH SDK · voices · MCP presentation and recovery | | [v2.3.0](./releases/2.3.0) | 2026-09-20 | User-configured managed JSON field acceptance and immutable artifact versions · Observe-mode execution evidence and controlled worker intervention · Approval/queued-input recovery and identity isolation · Skill document/folder uploads · DSH, reasoning-model, and SSE fixes | | [v2.2.0](./releases/2.2.0) | 2026-08-29 | **Runtime mainline** — employee runtime contract + provider registry + session lifecycle decouple employees from the built-in reasoning loop; **DeepSeek Harness** joins through an authenticated JSON-RPC child process while tools remain governed by host workspace policy and Tool Guard, with install/configure/verify/enable management; **durable Goals recover across bounded segments and backend restarts** (database queue · supervisor · leases · attempts · accepted-input queue · provider backoff · evidence-gated checklist completion); **bidirectional A2A** (Agent Cards · message send/stream · task get/cancel · `call_a2a_agent` · SSRF/timeout/response caps); hardened Team checkpoints, board recovery, deliverable gates, worker attachments, and long-form output; optional OfficeCLI + ACP prompt timeout + tighter workspace boundaries | | [v2.1.0](./releases/2.1.0) | 2026-08-15 | **Unified Team Runs** — one request, task DAG, worker execution, final synthesis, and deliverables share a `runId`; Chat delivery / Agents observation / Teams governance consume one projection, while worker conversations stay out of the normal sidebar · **closed skill-evolution loop** (cross-session recurring-request mining · reflection · constrained auto-binding · curator governance handover · origin policy · snapshots and restore points, workspace-scoped and conservative by default) · **replayable reasoning** (live `` extraction · UI wall-clock duration · all/terminal display controls · plain-text trajectory export) · proactive channel push + targeted Cron delivery · context-window override/probe/catalog budgeting · progressive tool bridge + action completion policy · hardened browser refs/navigation/waits · WebChat/SSE/LLM stream cleanup and timeouts · Feishu execution progress · Qwen3-ASR HTTP · batch session deletion · date-partitioned files · 64-bit id and numeric tool-schema precision fixes | diff --git a/mateclaw-server/src/main/resources/docs/en/roadmap.md b/mateclaw-server/src/main/resources/docs/en/roadmap.md index 65d6fa81..a8c96700 100644 --- a/mateclaw-server/src/main/resources/docs/en/roadmap.md +++ b/mateclaw-server/src/main/resources/docs/en/roadmap.md @@ -1,5 +1,8 @@ # Roadmap +3.0.0-SNAPSHOT is in development: Boot 4, Spring AI 2 and AgentScope 2 execution, retaining tool governance, history contracts and recovery boundaries. See Releases for stable builds. [Architecture changelog](./changelog) + + > "People don't know what they want until you show it to them." > > This isn't a feature list. It's a manifesto about **how your AI assistant should exist.** diff --git a/mateclaw-server/src/main/resources/docs/en/teams.md b/mateclaw-server/src/main/resources/docs/en/teams.md index 810f1d87..be295f24 100644 --- a/mateclaw-server/src/main/resources/docs/en/teams.md +++ b/mateclaw-server/src/main/resources/docs/en/teams.md @@ -196,3 +196,7 @@ They coexist: a team member can still call `delegateToAgent` inside its own task - [Persistent Goals](./goals) — cross-turn follow-through for a single employee - [Workflow](./workflow) — deterministic step orchestration (use workflows when the process is fixed, teams when it must be decomposed on the spot) - [Security & Approval](./security) — how tool approval relates to team task approval + +## 3.0.0: guarded worker-result judgment + +Worker-result judgment uses typed decisions, output contracts and task identity checks. Domain transactions commit outcomes and state. SHADOW remains default and does not replace approvals or artifact acceptance. diff --git a/mateclaw-server/src/main/resources/docs/en/tools.md b/mateclaw-server/src/main/resources/docs/en/tools.md index ff9730b3..83d6e06f 100644 --- a/mateclaw-server/src/main/resources/docs/en/tools.md +++ b/mateclaw-server/src/main/resources/docs/en/tools.md @@ -143,8 +143,8 @@ Output: "2026-04-11T14:30:22" Web search via a **provider chain** — DuckDuckGo and SearXNG as keyless fallbacks, Serper and Tavily when you have keys. Configured in `Settings → System → Search Service` and takes effect without restart. ``` -Input: {"query": "Spring AI Alibaba latest version", "freshness": "month", "count": 5} -Output: "Spring AI Alibaba 1.1 was released..." +Input: {"query": "AgentScope Java latest version", "freshness": "month", "count": 5} +Output: "AgentScope Java 2.0 was released..." ``` Features: diff --git a/mateclaw-server/src/main/resources/docs/en/workspaces.md b/mateclaw-server/src/main/resources/docs/en/workspaces.md index 4421a7df..82e97288 100644 --- a/mateclaw-server/src/main/resources/docs/en/workspaces.md +++ b/mateclaw-server/src/main/resources/docs/en/workspaces.md @@ -315,3 +315,7 @@ curl -X PUT http://localhost:18088/api/v1/workspaces/1/members/42 \ - [Security & Approval](./security) — how workspace isolation interacts with Tool Guard and File Guard - [LLM Wiki](./wiki) — workspace-scoped knowledge bases - [Memory](./memory) — workspace memory files + +## Member file isolation (3.0.0) + +Personal files can be isolated by member identity with shared-file fallback. This is opt-in and does not change default sharing; workspace roles and tool permissions still apply. diff --git a/mateclaw-server/src/main/resources/docs/zh/agents.md b/mateclaw-server/src/main/resources/docs/zh/agents.md index 9ec7e95e..3c5da7af 100644 --- a/mateclaw-server/src/main/resources/docs/zh/agents.md +++ b/mateclaw-server/src/main/resources/docs/zh/agents.md @@ -508,3 +508,7 @@ logging: - [工作流](./workflow)（1.3.0+）——把多个数字员工 + 系统动作编排成一条业务流程 - [触发器](./triggers)（1.3.0+）——让事件自动启动工作流或员工对话 - [架构说明](./architecture)——原生 ReAct 与计划运行时深入 + +## 执行与工具边界（3.0.0） + +AgentScope 2 驱动原生 ReAct 和计划单步内部循环；工具实际执行仍交给 MateClaw 的审批、权限和审计链。显式关闭工具的配置保持有效，员工草稿 JSON 可位于推理块之后。 diff --git a/mateclaw-server/src/main/resources/docs/zh/changelog.md b/mateclaw-server/src/main/resources/docs/zh/changelog.md new file mode 100644 index 00000000..cc724d93 --- /dev/null +++ b/mateclaw-server/src/main/resources/docs/zh/changelog.md @@ -0,0 +1,57 @@ +# 3.0.0-SNAPSHOT：执行架构升级 + +开发快照，尚未发布正式版本。 + +## 自 2.3.0 以来的累计更新 + +3.0.0-SNAPSHOT 同时收录原 2.4.0 开发周期的更新；2.4.0 未单独作为正式发行版发布。以下按能力归类，避免把架构迁移前的功能遗漏。 + +### 工作空间与会话 + +- 可选成员文件隔离：共享工作空间中按成员身份解析个人文件，保留共享文件回退；默认共享策略与工作空间角色权限仍适用（#644）。 +- 持久渠道会话过期查询上下文清理，避免旧请求影响新任务。上传图片从会话存储解析，目录附件保留目录信息并校验上传边界。 +- 员工创建时兼容推理块后的 JSON；显式关闭工具的员工不再因通用能力被重新赋予工具（#648、#655）。 + +### 结构化决策与目标、计划、团队 + +- 新增有界、类型化的 Choice / Boolean / Score 决策管线及持久审计，支持 OFF / SHADOW / ACTIVE 和场景覆盖。默认 SHADOW，建议不能绕过领域硬约束。 +- Goal 续跑决策、计划员工路由、团队 worker 结果判断共用结构化决策基础；结果与状态在领域事务中提交，延迟结果经身份与 fence 校验。 +- 增强路由上下文、worker 输出契约、委派步骤结果校验和有界 Goal 恢复；记录已应用的后续执行抑制，保留原完成验收门。 + +### 模型与 DeepSeek Harness + +- DSH 迁移至 SDK 运行时，新增锁定版本与受控升级；失败健康检查会失效，损坏配置仍可编辑修复（#650）。 +- 模型探测检查已完成响应的有效性，不把只有连接或空响应误判为成功（#658）。 + +### 朗读、聊天与工具展示 + +- 恢复 Edge 和 DashScope 朗读，支持供应商音色列表、选择与试听（#646）。 +- MCP 结构化结果支持专用工具展示，并放在最终回答下方；普通工具结果不再自动混进最终回答正文（#651、#656、#661）。 + +### 渠道与数据运维 + +- 飞书 WebSocket readiness 与资源生命周期由应用管理，关闭和重连释放所属资源（#659）。 +- 不支持的 webhook 渠道在写入配置前拒绝，避免保存无法运行的配置（#654）。 +- Flyway 校验失败默认拒绝启动，历史修复必须显式选择，保留备份与恢复指导（#657）。 +- 增加商业版概览与入口；这不意味着开源版包含商业模块。 + +## 平台与执行核心 + +Java 21，Spring Boot 4.1.1，Spring AI 2.0.1，AgentScope Java 2.0.4。移除 Spring AI Alibaba starter 和 Graph 依赖。AgentScope 驱动普通 ReAct 和 Plan-Execute 单步内部的推理、工具、观察循环；整体计划调度、Goal 预算与验收、团队协作仍由 MateClaw 管理。 + +工具批次经 ApplicationGuardedToolkit 转交应用执行器，继续执行审批、工作空间权限、审计与取消策略。模型请求走现有 Spring AI/provider 适配；应用继续负责 SSE、历史记录及持久化。 + +## 模型、工具与恢复 + +- 原生 DashScope 聊天与 Embedding 适配替代 Alibaba AI 集成，保留原生语音 SDK。 +- OpenAI/Anthropic 适配保留推理、媒体、缓存和用量；取消先终止所属 HTTP 调用，取消后不继续备用模型。 +- MCP SDK 0.18.3：保留调用 ID、结构化结果和协议错误，应用内部执行能力不写入网络元数据。 +- Goal 工具检查点在副作用前持久化执行身份；结果不确定时阻止盲目重放，保留审批恢复和会话隔离。 +- MyBatis Plus 3.5.17 使用 Spring Boot 4 starter；保留 JSON 字符串 ID 与已有会话数据契约。 +- 技能包二进制文件处理和内置文档镜像同步加固。 + +## 升级与验证 + +升级前停止旧进程并备份数据库、workspace、插件及配置。使用 Java 21 构建，检查第三方插件与新版依赖的兼容性；恢复旧版本时使用升级前备份，不能把升级后的数据库降级兼容视为已验证。 + +本轮完成 Maven reactor、前端构建、PostgreSQL/MySQL 迁移与事务分页、工具审批/取消、进程崩溃恢复及六小时持续运行验证。Kingbase 实机验证尚未覆盖。前端交互接口保持兼容，本轮没有重写聊天 UI。 diff --git a/mateclaw-server/src/main/resources/docs/zh/channels.md b/mateclaw-server/src/main/resources/docs/zh/channels.md index dfcbcc27..50e8b254 100644 --- a/mateclaw-server/src/main/resources/docs/zh/channels.md +++ b/mateclaw-server/src/main/resources/docs/zh/channels.md @@ -716,3 +716,7 @@ Cron 编辑器会持久保存 delivery channel 和 target；修改表达式或 - [聊天与消息](./chat)——消息流、segment、流式事件 - [Agent 引擎](./agents)——实际在每个渠道里回答的是什么 - [配置说明](./config)——全局渠道调优 + +## 渠道配置与生命周期（3.0.0） + +飞书 WebSocket readiness、重连与关闭由应用管理。无法运行的 webhook 渠道类型在保存前拒绝；持久会话中的过期查询上下文会清理，防止串入后续任务。 diff --git a/mateclaw-server/src/main/resources/docs/zh/chat.md b/mateclaw-server/src/main/resources/docs/zh/chat.md index 5bcf856a..cdbb8a5d 100644 --- a/mateclaw-server/src/main/resources/docs/zh/chat.md +++ b/mateclaw-server/src/main/resources/docs/zh/chat.md @@ -416,3 +416,7 @@ Sessions 页一次可选择最多 200 个会话批量删除。服务端会对每 - [记忆系统](./memory)——跨会话保留下来的是什么 - [多渠道接入](./channels)——聊天还能在哪里发生 - [LLM Wiki](./wiki)——Agent 在回答时能读到什么 + +## 结构化工具结果（3.0.0） + +MCP 结构化结果与最终回答分别展示，工具结果放在回答下方。普通工具结果不会自动拼入最终回答正文；工具调用 ID 和结构化内容保留在流式事件与历史中。图片附件从会话存储解析，目录附件保留目录信息。 diff --git a/mateclaw-server/src/main/resources/docs/zh/goals.md b/mateclaw-server/src/main/resources/docs/zh/goals.md index ef30f059..a88944fd 100644 --- a/mateclaw-server/src/main/resources/docs/zh/goals.md +++ b/mateclaw-server/src/main/resources/docs/zh/goals.md @@ -444,3 +444,7 @@ provider 全局退避只保存在当前后端实例内；重启恢复仍以数 **Goal 不是给员工加一个功能。是改它的状态。** 以前的员工"答完就忘"。Goal 让员工跨多轮记住一件事 — 它在干什么、还差什么、什么时候算完。你只用说一次。剩下的，让头像旁边那圈光替你跟。 + +## 3.0.0：续跑决策与副作用恢复 + +续跑建议经有界结构化决策与业务约束检查，默认 SHADOW。工具执行前保存带 fence 的检查点；副作用后结果不确定时要求干预而不是盲目重放。详情见[决策层](./decision-layer)和[更新日志](./changelog)。 diff --git a/mateclaw-server/src/main/resources/docs/zh/index.md b/mateclaw-server/src/main/resources/docs/zh/index.md index 05bfa899..c601dcad 100644 --- a/mateclaw-server/src/main/resources/docs/zh/index.md +++ b/mateclaw-server/src/main/resources/docs/zh/index.md @@ -19,10 +19,13 @@ hero: text: GitHub link: https://github.com/mateaix/mateclaw - theme: alt - text: v2.3.0 更新日志 - link: /zh/releases/2.3.0 + text: 3.0.0 快照更新日志 + link: /zh/releases/3.0.0 features: + - icon: ⚙️ + title: 3.0.0-SNAPSHOT — 执行架构升级 + details: 3.0.0-SNAPSHOT 升级到 Java 21、Spring Boot 4 和 Spring AI 2，AgentScope 2 驱动 ReAct 与计划单步执行循环。工具审批、权限、Goal、SSE 与持久化继续由 MateClaw 统一治理；这是开发快照。 - icon: ✅ title: 2.3.0 — 托管 JSON 验收 details: 用户指定顶层必需字段，服务端把检查绑定到当前要求与不可变 JSON 版本。默认观察模式的执行证据用于追踪；普通产物诊断不代替验收绑定。 diff --git a/mateclaw-server/src/main/resources/docs/zh/intro.md b/mateclaw-server/src/main/resources/docs/zh/intro.md index c006c83e..a5178b11 100644 --- a/mateclaw-server/src/main/resources/docs/zh/intro.md +++ b/mateclaw-server/src/main/resources/docs/zh/intro.md @@ -7,6 +7,11 @@ head: content: MateClaw,多智能体,AI操作系统,自部署AI,Spring AI + AgentScope,ReAct,Plan-and-Execute,MCP,LLM Wiki,记忆系统,Tool Guard,开源 --- +3.0.0-SNAPSHOT 升级到 Java 21、Spring Boot 4 和 Spring AI 2，AgentScope 2 驱动 ReAct 与计划单步执行循环。工具审批、权限、Goal、SSE 与持久化继续由 MateClaw 统一治理；这是开发快照。 + +[升级说明与 changelog](./changelog) + + # MateClaw — 自部署多智能体 AI 操作系统 **v2.3.0（2026-09-20 发布）——持久目标的 JSON 验收与执行证据。** 用户可配置 JSON 必需字段，将检查结果绑定到当前要求和不可变产物版本；同时加入团队成员受控干预、技能文档及文件夹上传，并修复 DSH 历史与输出预算、vLLM 推理内容和 SSE 分帧。 [v2.3.0](./releases/2.3.0) diff --git a/mateclaw-server/src/main/resources/docs/zh/mcp.md b/mateclaw-server/src/main/resources/docs/zh/mcp.md index bc968638..8fadf08d 100644 --- a/mateclaw-server/src/main/resources/docs/zh/mcp.md +++ b/mateclaw-server/src/main/resources/docs/zh/mcp.md @@ -614,3 +614,7 @@ registerToolResultRenderer('inventory', { ``` 组件通过 `data` prop 接收数据。工具随后可以在 `blocks` 中返回 `{ "type": "inventory", "data": { "name": "示例", "quantity": 10 } }`。注册新组件需要重新构建前端；工具返回值只能选择已注册类型，不能导入组件或执行脚本。自定义组件需校验自身数据结构，并安全处理不可信内容。 + +## 3.0.0：协议与展示边界 + +SDK 升级至 0.18.3。调用 ID、结构化内容和协议错误保留；普通工具结果在最终回答下方独立展示。传输失败且副作用不确定时不自动重复执行。 diff --git a/mateclaw-server/src/main/resources/docs/zh/models.md b/mateclaw-server/src/main/resources/docs/zh/models.md index 376f15c7..fdf95545 100644 --- a/mateclaw-server/src/main/resources/docs/zh/models.md +++ b/mateclaw-server/src/main/resources/docs/zh/models.md @@ -527,3 +527,7 @@ OpenAI 兼容链路同时保留工具 JSON Schema 中的 `integer` / `number`。 - [配置说明](./config)——完整配置参考 - [Agent 引擎](./agents)——Agent 怎么使用模型 - [控制台](./console)——模型管理 UI + +## 模型探测与取消（3.0.0） + +探测必须拿到有效的完成响应，连接成功或空响应不能代表模型可用。停止生成会取消本次请求所属的 HTTP 调用，停止后不再切换备用模型。 diff --git a/mateclaw-server/src/main/resources/docs/zh/multimodal.md b/mateclaw-server/src/main/resources/docs/zh/multimodal.md index 64de6c2e..33c082bf 100644 --- a/mateclaw-server/src/main/resources/docs/zh/multimodal.md +++ b/mateclaw-server/src/main/resources/docs/zh/multimodal.md @@ -197,3 +197,7 @@ Agent 调用它们和调用任何其他工具一样。工具层负责供应商 - [聊天与消息](./chat)——附件输入、多模态旁路路由、生成的媒体如何挂载到消息上 - [模型配置](./models)——供应商配置 UI、多模态旁路设置 - [工具系统](./tools)——承载多模态生成的工具层 + +## 朗读音色与试听（3.0.0） + +聊天朗读支持 Edge 与 DashScope。音色按供应商提供列表，可选择并试听；可用音色取决于供应商配置。接口为 GET /api/v1/tts/voices、POST /api/v1/tts/preview 和 POST /api/v1/tts/synthesize。 diff --git a/mateclaw-server/src/main/resources/docs/zh/quickstart.md b/mateclaw-server/src/main/resources/docs/zh/quickstart.md index 3f06da8c..e1070da0 100644 --- a/mateclaw-server/src/main/resources/docs/zh/quickstart.md +++ b/mateclaw-server/src/main/resources/docs/zh/quickstart.md @@ -1,5 +1,8 @@ # 快速开始 +源码开发分支当前为 `3.0.0-SNAPSHOT`；正式安装包仍以 Releases 中已发布的版本为准。升级前请阅读[升级说明](./changelog#升级与验证)。 + + 60 秒到第一条消息。**只有一条路：桌面端。** Docker 和源码启动在 [配置说明](./config) 和 [贡献指南](./contributing) 里。这一页只做一件事——用尽可能快的速度，把你从"什么都没有"送到"一个能工作的 Agent"面前。 diff --git a/mateclaw-server/src/main/resources/docs/zh/releases.md b/mateclaw-server/src/main/resources/docs/zh/releases.md index 712d3f6c..820aab48 100644 --- a/mateclaw-server/src/main/resources/docs/zh/releases.md +++ b/mateclaw-server/src/main/resources/docs/zh/releases.md @@ -10,6 +10,7 @@ | 版本 | 日期 | 亮点 | |------|------|------| +| [3.0.0-SNAPSHOT](./releases/3.0.0) | 开发中 | Java 21 · Boot 4 · Spring AI 2 · AgentScope 2 执行循环 · 结构化决策 · 成员文件隔离 · DSH SDK · 朗读音色 · MCP 展示与恢复 | | [v2.3.0](./releases/2.3.0) | 2026-09-20 | 用户配置的托管 JSON 字段验收与不可变产物版本 · 观察模式执行证据与受控团队干预 · 审批/排队输入恢复与身份隔离 · 技能文档/文件夹上传 · DSH、推理模型与 SSE 加固 | | [v2.2.0](./releases/2.2.0) | 2026-08-29 | **Runtime 主线**——员工运行时 contract + provider registry + session 生命周期，把员工与内置推理循环解耦；**DeepSeek Harness** 以认证 JSON-RPC 子进程接入，工具继续经过宿主工作空间策略与 Tool Guard，并提供安装/配置/校验/启停管理；**持久目标跨有界片段和后端重启恢复**（数据库队列 · supervisor · lease · attempt · 输入排队 · provider 退避 · checklist 证据完成门）；**A2A 双向互联**（Agent Card · message/send/stream · task 查询/取消 · `call_a2a_agent` · SSRF/超时/响应上限）；Team checkpoint / 任务板恢复 / 交付物完成门 / worker 附件与长回答加固；可选 OfficeCLI 引擎 + ACP prompt timeout + 工作空间边界收口 | | [v2.1.0](./releases/2.1.0) | 2026-08-15 | **统一 Team Run**——一次团队请求、任务 DAG、成员执行、最终汇总与交付物共用 `runId`，Chat 成果交付 / Agents 实时观察 / Teams 历史治理读取同一投影，成员子会话不再污染普通会话列表 · **Skill 自进化闭环**（跨会话重复请求 mining · reflection · 受约束自动绑定 · curator 治理移交 · origin 策略 · 快照与恢复点，按工作空间隔离且默认保守） · **推理轨迹可回放**（实时 `` 提取 · UI 真实耗时 · 全部/最终轮次控制 · 纯文本 trajectory 导出） · 主动渠道消息 + Cron 定向投递 · 模型窗口覆盖/探测/目录预算 · 渐进式工具桥 + 行动完成约束 · 浏览器 ref/导航/等待加固 · WebChat/SSE/LLM 流回收与超时 · 飞书执行进度 · Qwen3-ASR HTTP · 会话批量删除 · 按日文件目录 · 64 位 id 与数值工具 schema 精度修复 | diff --git a/mateclaw-server/src/main/resources/docs/zh/roadmap.md b/mateclaw-server/src/main/resources/docs/zh/roadmap.md index 5558927b..5d8a1840 100644 --- a/mateclaw-server/src/main/resources/docs/zh/roadmap.md +++ b/mateclaw-server/src/main/resources/docs/zh/roadmap.md @@ -1,5 +1,8 @@ # 路线图 +3.0.0-SNAPSHOT 开发中：Boot 4、Spring AI 2 与 AgentScope 2 执行架构升级，保留工具治理、历史接口和恢复边界。正式发行版以 Releases 为准。 [架构升级日志](./changelog) + + > "人们不知道自己想要什么，直到你把它摆在他们面前。" > > 这不是一份功能清单。这是一个关于**你的 AI 助手应该如何存在**的宣言。 diff --git a/mateclaw-server/src/main/resources/docs/zh/teams.md b/mateclaw-server/src/main/resources/docs/zh/teams.md index af30cc1e..92621b7b 100644 --- a/mateclaw-server/src/main/resources/docs/zh/teams.md +++ b/mateclaw-server/src/main/resources/docs/zh/teams.md @@ -196,3 +196,7 @@ Lead 不限定 Agent 类型。**ReAct 型 Lead** 用 `team_tasks` 逐条建任 - [持久化目标](./goals)——单个员工的跨轮任务跟进 - [工作流](./workflow)——确定性步骤编排（流程固定时用工作流，流程要临场拆解时用团队） - [安全与审批](./security)——工具审批与团队任务审批的关系 + +## 3.0.0：受控 worker 结果判断 + +worker 结果判断经类型化决策管线，校验输出契约与任务身份，结果和状态通过领域事务提交。默认 SHADOW，不替代原有审批或交付物验收。 diff --git a/mateclaw-server/src/main/resources/docs/zh/tools.md b/mateclaw-server/src/main/resources/docs/zh/tools.md index 13162340..64655e22 100644 --- a/mateclaw-server/src/main/resources/docs/zh/tools.md +++ b/mateclaw-server/src/main/resources/docs/zh/tools.md @@ -143,8 +143,8 @@ Tool Guard 是守门员。超时是**每个工具独立**的（这样一个慢 通过**供应商链**搜索——DuckDuckGo 和 SearXNG 作为无 Key 的 fallback，Serper 和 Tavily 在你有 Key 时启用。在 `设置 → 系统设置 → 搜索服务` 里配置，**改完即生效，不需要重启**。 ``` -输入：{"query": "Spring AI Alibaba 最新版本", "freshness": "month", "count": 5} -输出："Spring AI Alibaba 1.1 发布..." +输入：{"query": "AgentScope Java 最新版本", "freshness": "month", "count": 5} +输出："AgentScope Java 2.0 发布..." ``` 特性： diff --git a/mateclaw-server/src/main/resources/docs/zh/workspaces.md b/mateclaw-server/src/main/resources/docs/zh/workspaces.md index 5f58fbf9..483eb3e1 100644 --- a/mateclaw-server/src/main/resources/docs/zh/workspaces.md +++ b/mateclaw-server/src/main/resources/docs/zh/workspaces.md @@ -320,3 +320,7 @@ curl -X PUT http://localhost:18088/api/v1/workspaces/1/members/42 \ - [安全与审批](./security)——工作空间隔离和 Tool Guard、File Guard 的交互 - [LLM Wiki](./wiki)——工作空间 scope 的知识库 - [记忆系统](./memory)——工作空间记忆文件 + +## 成员文件隔离（3.0.0） + +可按成员身份隔离个人文件，保留共享文件回退。此能力是可选配置，不改变默认共享模式；个人文件访问仍受工作空间角色与工具权限控制。 diff --git a/mateclaw-ui/package.json b/mateclaw-ui/package.json index 9d84d731..118a6c45 100644 --- a/mateclaw-ui/package.json +++ b/mateclaw-ui/package.json @@ -1,6 +1,6 @@ { "name": "mateclaw-ui", - "version": "2.4.0-SNAPSHOT", + "version": "3.0.0-SNAPSHOT", "private": true, "type": "module", "description": "MateClaw - Personal AI Assistant Web Console", diff --git a/mateclaw-ui/src/views/Settings/About/index.vue b/mateclaw-ui/src/views/Settings/About/index.vue index 305a1d56..3f8fbb6d 100644 --- a/mateclaw-ui/src/views/Settings/About/index.vue +++ b/mateclaw-ui/src/views/Settings/About/index.vue @@ -57,12 +57,12 @@ const pillars = computed(() => [ ]) const techStack = [ - { icon: markRaw(Coffee), name: 'Spring Boot', version: '3.3' }, - { icon: markRaw(Cpu), name: 'Spring AI', version: '1.1' }, - { icon: markRaw(Promotion), name: 'Spring AI Alibaba', version: '1.1' }, + { icon: markRaw(Coffee), name: 'Spring Boot', version: '4.1.1' }, + { icon: markRaw(Cpu), name: 'Spring AI', version: '2.0.1' }, + { icon: markRaw(Promotion), name: 'AgentScope', version: '2.0.4' }, { icon: markRaw(Monitor), name: 'Vue 3', version: '3.5' }, { icon: markRaw(Lightning), name: 'Vite', version: '7' }, - { icon: markRaw(Coin), name: 'MyBatis Plus', version: '3.5' }, + { icon: markRaw(Coin), name: 'MyBatis Plus', version: '3.5.17' }, ] diff --git a/pom.xml b/pom.xml index b8a7153d..91a3646c 100644 --- a/pom.xml +++ b/pom.xml @@ -23,7 +23,7 @@ - 2.4.0-SNAPSHOT + 3.0.0-SNAPSHOT 21