Releases: lujoai/Lujo-MCP
Releases · lujoai/Lujo-MCP
Release list
Lujo-MCP v0.5.0
Highlights
- DebugContext Schema Alignment — Pydantic model expanded from 7 to 20 fields, aligned with
build_debug_context()output - Runtime Debug Context Integration —
build_debug_context()now returnsDebugContext | None(Pydantic model) instead of plain dict - MCP Tool Category Metadata — All 17 MCP tools tagged with
category(agent/sdk) andexperimentalfields intools/listresponse - Prompt Injection Protection — LLM analyzer and Agent layer hardened with evidence wrapping (
<debug_evidence>XML tags) and injection guard - API Schema Validation —
/verifyand/verify/uiendpoints now use Pydantic models (VerifyRequest/VerifyUiRequest) withextra="ignore" - Session Security Hardening — Session limit (10,000) with LRU eviction,
/internal/healthIP-based access control
Compatibility
- MCP Protocol unchanged (
2024-11-05) - Existing MCP clients compatible (new
category/experimentalfields are optional, ignored by old clients) - No breaking changes — all new fields are Optional with defaults
- DebugContext JSON structure unchanged (
model_dump()produces equivalent output)
Not included
The following features are not implemented in v0.5.0:
- Repair Loop — planned for v0.5.2
- Auto Patch — planned for v0.5.2
- Automatic Code Repair — planned for v0.5.2
- Source Map parsing — planned for v0.5.1
- Browser SDK enhancements — planned for v0.5.1
Test Results
992 passed, 6 skipped, 0 failed
Lujo-MCP v0.4.1-beta
Lujo-MCP 是面向 AI Agent 的运行时调试上下文基础设施。
目标:让 AI Agent 不仅读取代码,还能够理解真实 Bug 运行现场。
What's New
Runtime Debug Context
- 采集:runtime events、exceptions、stacktrace、network trace、UI events、git context、spec context
Debug Experience RAG
- 通过历史 Debug Experience 增强 Agent 分析能力
retrieve_debug_experience()三层检索:L1 fingerprint 精确 → L2 message normalize → L3 vector(默认关闭)- 种子知识库 + KB 验证写回
MCP Debug Context Observability(Phase 3 D5)
DebugContextTrace:记录 request_id / Runtime Context 可用性与大小 / Debug Experience 开关与命中数 / Context 构建耗时 / Tool 响应耗时- stdio + HTTP 传输层记录 tool 响应耗时/大小(仅日志,不打印敏感负载)
Server(...)传入version=0.4.1-beta对齐 serverInfo 版本
Quality Benchmark Framework(Phase 3 D6)
benchmark/:BenchmarkCase/EvaluationMetrics+ 5 个手写 Debug Case(api_500 / frontend_blank / db_error / auth_403 / perf_slow)runner.pyCLI:list / show / quality 旁证- 验证 MCP Debug Context 是否提升外部 AI Debug 能力(与 QualityScorer 两个体系分离)
npm 开箱即用分发
npm install -g @lujoai/lujo-mcp一条命令,三平台二进制(win32-x64 / linux-x64 / osx-arm64)自动安装- 版本统一
0.4.1-beta(app / npm / README / CHANGELOG / serverInfo / git tag)
Architecture
External MCP Client
↓
MCP Server(stdio / HTTP)
↓
Runtime Debug Context
↓
Debug Experience RAG
↓
Agent
↓
Verifier
Validation
测试:927 passed / 6 skipped / 0 failed(含 D5 MCP Observability 16 项 + D6 Benchmark 框架 19 项 + stacktrace 工具与存储工厂边界 17 项)
MCP Client 接入验证:stdio + HTTP 双传输 initialize / tools/list(17 工具)/ tools/call 全通过。
说明:无回归。
Architecture Rules
- Runtime 保持纯净(不依赖 RAG / Agent / LLM / MCP)
- RAG 独立(不反向依赖 Agent / Runtime / LLM / MCP)
- Agent 消费 RAG(允许方向:Agent → RAG)
Limitations
当前版本:
- 不包含自动修复
- 不包含 Patch 生成
- 不包含 Repair Loop
Roadmap
未来:
- More Debug Experience
- Better Context Collection
- Enterprise Integration
Lujo-MCP v0.4.0
正式版 v0.4.0 — AI 智能调试,开箱即用
🏆 最大亮点:npm 一键安装
告别 Python 环境配置,小白用户也能直接用:
npm install -g @lujoai/lujo-mcp✨ 核心能力
- Runtime Debug Context(真实运行现场采集)
- Debug Experience RAG(历史调试经验检索增强)
- AI Debug Agent(自动修复 + 多 Agent DAG 协同)
- Browser SDK V2-V6(网络请求捕获 / 静默失败检测)
- 矢量检索 RAG(in-process + Qdrant 双后端)
✅ 质量验证
- 单元测试:891 passed / 6 skipped / 0 failed
- CI 构建:Windows/Linux/macOS 三平台全通过
- npm 发布:3 平台二进制 + 元包全部上线
🔧 MCP 客户端配置示例
{
"mcpServers": {
"lujo-mcp": {
"command": "lujo-mcp-server"
}
}
}📝 变更摘要
相比 v0.3.0,本版本新增:
- MCP Streamable HTTP 传输
- AI Debug Agent Phase 1 + Phase 2
- Dashboard SSE 实时推送
- 安全加固(RBAC/LFI/SSRF 防护)
- npm 跨平台分发
Lujo-MCP v0.4.0-beta
Lujo-MCP v0.4.0-beta
Overview
Lujo-MCP 是面向 AI Agent 的运行时调试上下文基础设施。
目标:让 AI Agent 不仅读取代码,还能够理解真实 Bug 运行现场。
What's New
Runtime Debug Context
采集:
- runtime events
- exceptions
- logs
- user behavior context
Debug Experience RAG
通过历史 Debug Experience 增强 Agent 分析能力:
DebugExperienceRecord输出 DTO/View(纯 View,不建存储、不替代 DebugCase)retrieve_debug_experience()三层检索:L1 fingerprint 精确 → L2 message normalize → L3 vector(默认关闭)
Context Assembly
Agent 上下文构建与 RAG 解耦:
_safe_debug_experience_recall()集成于assemble()输出(可选字段,默认 None)debug_experience_enabled默认False,关闭状态零调用、零耗时
Verification Foundation
提供验证能力基础(Verifier / assert engine / spec store)。
Architecture
Runtime
↓
Debug Context
↓
Debug Experience RAG
↓
Agent
↓
Verifier
Validation
测试:874 passed / 6 skipped / 0 failed
说明:无回归。
Architecture Rules
- Runtime 保持纯净(不依赖 RAG / Agent / LLM / MCP)
- RAG 独立(不反向依赖 Agent / Runtime / LLM / MCP)
- Agent 消费 RAG(允许方向:Agent → RAG)
Limitations
当前版本:
- 不包含自动修复
- 不包含 Patch 生成
- 不包含 Repair Loop
Roadmap
未来:
- Debug Experience Quality Evaluation
- Benchmark Dataset
- Repair Loop Research
- Human Approval Workflow