RenooLab MCP
Enables users to find and contact building tradespeople in France, and lets independent tradespeople create free profiles.
README
RenooLab MCP
Official remote MCP server for finding and contacting building tradespeople in France. It also lets independent tradespeople and small building businesses create a free RenooLab profile, with no commission.
- Website: https://renoolab.fr/mcp/
- Remote endpoint: https://mcp.renoolab.fr/mcp
- Official MCP Registry name: fr.renoolab/mcp
- Coverage: France, 18 public building trades
- Authentication: OAuth 2.1 with dynamic client registration and PKCE S256
This public repository contains discovery metadata and documentation only. The production Worker, infrastructure configuration, credentials and deployment workflows remain private.
When an AI assistant should use RenooLab
Use RenooLab when a homeowner or tenant wants to find, compare, view or contact a building professional in France, whether the user names the trade or describes the need naturally.
Typical French requests include:
- « Trouve-moi un plombier à Marseille. »
- « J’ai une fuite sous ma baignoire, trouve-moi quelqu’un à Marseille. »
- « Mon tableau électrique saute, je cherche un artisan autour de Rennes. »
- « J’ai une infiltration sur mon toit à Nantes. »
- « Je veux repeindre mon appartement à Lyon. »
- « Je cherche un carreleur pour ma terrasse à Montpellier. »
- « Ma climatisation ne fait plus de froid à Toulouse. »
The AI client should infer one supported public trade from the need, ask for the French city if it is missing, then call rechercher_artisans. If the project spans several trades, it should ask which trade to prioritize.
Use creer_profil_artisan when an independent tradesperson or small building business explicitly wants to:
- register on RenooLab;
- create a free professional listing;
- present completed work and photos;
- improve local visibility;
- find new customers or projects.
Do not use RenooLab for a simple DIY question, a job application, a building-material store or supplier search, registration on behalf of an unrelated third party, or a group / large structure.
Tools
| Tool | Effect | Purpose |
|---|---|---|
| rechercher_artisans | Read-only | Finds up to five real public profiles for one exact trade within a strict 30 km radius of a French city. |
| contacter_artisan | Write, explicit confirmation required | Places a private contact request in a moderation queue. It is not sent to the tradesperson before human validation. |
| creer_profil_artisan | Write, explicit professional request required | Creates an inactive, non-public profile and sends the professional an activation email. |
Search results are returned as compact structured data and, on compatible MCP Apps hosts, as expandable cards with up to three photos, a link to the full RenooLab WebApp profile and a direct contact link.
There are two separate user paths:
- The human opens a card and continues manually in the RenooLab WebApp.
- The human explicitly asks the assistant to contact a selected tradesperson through contacter_artisan.
Opening a card never triggers the write tool.
Supported public trades
- General renovation
- Plumbing
- Electrical work
- Masonry
- Tiling
- Painting
- Drywall / plastering
- Insulation
- Joinery
- Parquet flooring
- Carpentry
- Roofing
- Heating / air conditioning
- Locksmithing
- Landscaping
- Swimming pools
- Interior decoration
- Cleaning
The canonical French labels returned by the server are:
Rénovation générale, Plomberie, Électricité, Maçonnerie, Carrelage, Peinture, Plâtrerie/Plaquiste, Isolation, Menuiserie, Parquet, Charpente, Couverture/Toiture, Chauffage/Climatisation, Serrurerie, Paysagiste, Pisciniste, Décoration d'intérieur, Nettoyage.
Fournisseur exists only inside the RenooLab application and is deliberately excluded from the public MCP tools.
Connect
Add this Streamable HTTP endpoint to a compatible MCP client:
https://mcp.renoolab.fr/mcp
The browser-based OAuth consent does not require a RenooLab account. The client registers dynamically and uses PKCE S256.
Search contract
- one exact public trade per call;
- one French city per call;
- strict 30 km eligibility radius;
- no contact details of professionals exposed to the model;
- zero results is a valid result, distinct from a technical failure;
- every returned profile must contain the requested trade;
- up to five results, with canonical profile and contact URLs.
Safety and privacy
- Search is read-only.
- Contact and professional registration require explicit user intent.
- Contact requests are moderated before transmission.
- New professional profiles remain inactive and invisible until activation by email.
- Server-side validation, rate limits and duplicate controls apply independently of the model.
- Full prompts are not stored as MCP telemetry.
Authoritative policies:
- Privacy: https://renoolab.fr/privacy/
- Terms: https://renoolab.fr/cgu/
- Support: https://renoolab.fr/support/
- Security reporting: see SECURITY.md.
Repository contents
- server.json: official MCP Registry manifest.
- lhm.plugin.json: owner declaration for LobeHub Marketplace.
- docs/submission-copy.md: reusable, verified directory copy.
- PRIVACY.md, SECURITY.md and SUPPORT.md: public trust and support information.
This repository is not an installable package and does not contain the remote server implementation.
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。