一个基于 Laravel 13 + Livewire 4 的演示应用,实现了集成 CRM(Pipedrive)的 AI 销售代理,包含队列驱动的 CRM 镜像、魔法链接客户端认证以及流式代理对话。

Stars

13

7 天增长

暂无数据

Fork 数

19

开放 Issue

0

开源协议

暂无数据

最近更新

2026-07-25

AI 仓库情报摘要
FR-AI / ANALYSIS

为什么值得关注

它展示了使用现代 PHP 栈构建 AI 销售代理的完整方案,集成了 LLM 可观测性(Langfuse)和长期记忆(Mem0),为开发者提供了实用参考。

适合谁使用

  • 探索 AI 代理集成的 Laravel 开发者
  • 构建 AI 驱动的销售或 CRM 工具的开发者
  • 学习流式代理和记忆模式的团队
  • 跟随配套视频课程的学习者或讲师

典型使用场景

  • 原型化一个利用 CRM 数据处理客户咨询的 AI 销售代理
  • 学习将 CRM 管道和阶段镜像到本地模型以提供 AI 上下文
  • 在 Laravel 应用中演示代理可观测性和记忆功能
  • 用作 AI 代理开发工作坊或课程的教学材料

项目优势

  • 使用 Beer and Code Harness 实现模块化代理设计(技能、钩子、命令)
  • 通过 Langfuse 内置 LLM 可观测性,支持追踪、成本评估和评价
  • 集成 Mem0 提供长期记忆,确保代理上下文持久化
  • 采用现代 Laravel 栈(13、Livewire 4、Sail、Pest、PHPStan),保证代码质量

使用前须知

  • 仅支持 Pipedrive 作为 CRM 驱动,其他 CRM 需要自定义开发
  • 需要外部 API 密钥(Langfuse、Mem0)和基础设施(Docker、队列)
  • 设计为实验室/演示项目,未针对生产环境加固(例如仅依赖魔法链接认证)

README 快速开始

Lab Agent Seller

Aplicação Laravel 13 + Livewire 4 que demonstra um agente de vendas com IA. Empresas conectam um CRM (driver Pipedrive), um job em fila (ScanCrmConnection) espelha pipelines/stages/campos/pessoas/deals para modelos locais, clientes entram via magic link (guard auth:client) e conversam com o agente de vendas streaming SellerAgent (laravel/ai).

Repositórios e ferramentas usados

ProjetoDescriçãoLink
Beer and Code HarnessHarness de agentes (skills, hooks, comandos e guidelines) usado no projetohttps://github.com/beerandcodeteam/beer-and-code-harness
LangfuseObservabilidade de LLM (traces, custos, avaliações)https://github.com/langfuse/langfuse
Mem0Camada de memória de longo prazo para o agentehttps://github.com/mem0ai/mem0

Pacotes de integração:

  • Langfuse — via axyr/laravel-langfuse (região US, projeto ai-lab).
  • Mem0 — via API HTTP (MEM0_API_KEY, MEM0_BASE_URL, MEM0_TIMEOUT no .env).

Aulas

Slides gravados neste repositório:

  • Aula 1slides-da-aula/aula-dia-1/index.html
  • Aula 2slides-da-aula/aula-dia-2/aula.html

Abra os arquivos .html direto no navegador.

Stack

  • PHP 8.5 · Laravel 13 · Livewire 4
  • laravel/ai (agente streaming)
  • Laravel Sail (Docker) · PostgreSQL 18 · Redis · Mailpit
  • Pest 4 · Larastan (PHPStan nível 7) · Pint
  • Tailwind CSS v4 · Vite

Setup do ambiente

Pré-requisitos: Docker, Composer 2, Node 22 + npm.

# 1. Clonar
git clone git@github.com:beerandcodeteam/lab-agent-seller.git
cd lab-agent-seller

# 2. Instalar deps PHP (necessário pra ter o binário do Sail)
composer install

# 3. Subir os containers (laravel.test PHP 8.5, pgsql 18, redis, mailpit)
vendor/bin/sail up -d

# 4. Setup completo: copia .env, key:generate, migrate --force, npm install, npm run build
vendor/bin/sail composer setup

# 5. Configurar chaves de API no .env
#    MEM0_API_KEY=...          (Mem0)
#    LANGFUSE_PUBLIC_KEY=...    (Langfuse - região US)
#    LANGFUSE_SECRET_KEY=...

Levantar o projeto

# Sobe os containers (se ainda não estiverem no ar)
vendor/bin/sail up -d

# Inicia os processos de desenvolvimento (servidor, queue, vite, logs)
vendor/bin

相关仓库与替代方案

根据分类、Topic 和编程语言匹配的相似项目。

ddosi
精选
ddosi GitHub avatar

PacketLens

PacketLens is a pure front-end, offline pcap analysis tool that runs entirely in the browser, supporting deep protocol decoding, HTTPS decryption, and million-packet instant loading without any backend.

HTML
17
ViffyGwaanl
精选
ViffyGwaanl GitHub avatar

kimi-k3-learn

An interactive learning system that turns the 47-page Kimi K3 technical report into a single offline HTML file with running algorithms, 3D visualizations, spaced repetition quizzes, and a smart highlighting QA tool.

AI 与机器学习
15
iamtechartist
精选
iamtechartist GitHub avatar

human-cell-visualizer

An interactive 3D visualization of three human cell types using Three.js, WebGL particles, and custom GLSL shaders, allowing users to rotate, zoom, morph, and explore annotated structures.

设计与创意
13