Traceable
Traceable est un espace de travail documentaire qui maintient vos enregistrements de conception connectés et visibles. Il remplace les feuilles de calcul, les wikis et les outils déconnectés par un lieu unique pour rédiger, tracer, contrôler, réviser et publier la documentation exigée par vos normes, avec une traçabilité en direct, des signatures électroniques verrouillées et une sortie prête pour l'audit. Il gère votre vérité pour le matériel et les logiciels, en particulier les produits réglementés tels que les dispositifs médicaux, les SaMD, les IVD et la Défense.
Serveur MCP hébergé
npx add-mcp 'https://app.traceabledocs.com/api/mcp/mcp'S’installe dans Claude Code, Codex, Cursor et plus
Documentation
Traceable’s AI Connector (MCP) lets Claude, or another AI client that accepts a custom server URL, work inside your Traceable account under your own permissions. It can read and draft; publishing, review and e-signature stay human actions, and every change is recorded against your name. It is off until an organisation admin turns it on.
AI and integrations
Working with an AI assistant (MCP)
Connect Claude, ChatGPT or Cursor to your projects, and what it can and cannot do.
Updated 2026-10-03
MCP connector reference
The full connector manual: access, connecting from each client, the consent screen, and all 25 tools in detail.
Updated 2026-10-03
Guides
Design history file (DHF)
What a DHF contains under the QMSR, how it relates to the medical device file, and how to keep it audit-ready.
Traceability matrix for medical devices
From user needs to verification, and how to keep the matrix honest between audits.
ISO 13485 document control
What clauses 4.2.4 and 4.2.5 expect, and how to meet them without the manual overhead.
IEC 62304: a practical guide
Software safety classes, key deliverables, and the role of traceability.
ISO 14971 risk management
The risk process, and why risk traceability is the part that breaks.
Using AI in a validated documentation process
How to let your team use AI on controlled documents without losing control of them.
Free templates
Requirements traceability matrix template
A free RTM in Excel and CSV, with a worked example.
IEC 62304 checklist
The key life-cycle deliverables and activities, scaled to your software safety class.
All templates
Every free template for regulated product teams.
Resources
Brand assets
The Traceable logo, app icon and mark to download in SVG and PNG, each at a permanent URL, with brand colours and how to use the name.
Connecting an assistant with the AI Connector
Which AI assistants can connect to Traceable?
Claude on claude.ai, Claude Code, ChatGPT (developer mode) and Cursor are verified end to end. The connector is an ordinary remote MCP server, so any client that accepts a custom server URL can use it, including VS Code. Clients that only speak stdio can connect through npx mcp-remote.
Do I need anything switched on first?
Yes, two things, and both must be on. Your organisation must permit external agents, which it does by default, and your own user must be granted MCP access, which is off by default. An organisation admin turns that on. If you belong to more than one organisation, every one of them must permit it.
What can an assistant change in my account?
Drafts only. It can read projects and documents, author and edit content, build ID grids, create trace links, and search the Reference Library, all under your own permissions. Publishing, review, approval and e-signature stay human actions, and every change it makes is recorded in the audit trail against your name.
Can it see everything in my organisation?
No. It acts as you, through the same project and organisation membership as the web app. If you are not a member of a project, neither is the assistant. If your role is read-only, so is the assistant.
Is the connector included in my plan?
Yes, on every plan. It is included in Essentials, Team, Regulated and Enterprise, and available with rate limits on Free. See the pricing page for plans and prices.