From Claude Cowork to a governed, company-wide AI system.

For the full value, yes. Bridge is designed to manage a complete Cowork rollout: single admin console, MDM control, plugins, agents, and audit logs unified inside your own cloud. Technically, Bridge’s MCP plugins and Dify workflows can also be used with Claude Team or Claude Enterprise, since MCP is an open standard, but in that setup you lose the parts that make Bridge powerful: MDM management, in-tenant data residency, unified audit logs, and full GDPR/DORA defensibility. If you already run Claude another way, bring the setup to a deployment review and we will tell you honestly what Bridge can and cannot manage in your environment.

In your own cloud. Bridge is deployed as a single-tenant solution into the same AWS account and VPC as your Private Claude Cowork: one perimeter, one owner. Your data, prompts, files, audit logs, and vector stores never leave your AWS environment. Moterra operates the platform as your data processor, but does not host any of your data.

Through your existing identity provider. Bridge supports SSO via Microsoft Entra ID and Google Workspace out of the box. Users log in with the same corporate credentials they already use, with no separate Claude accounts to manage. Joiners and leavers flow through the same provisioning process as the rest of your estate.

Bridge ships with four standard MCP plugins that let Claude Cowork work directly with your business systems: SharePoint RAG (Claude answers questions grounded in your internal documents), Text-to-SQL (Claude queries your databases in plain language), Web Search & Scrape (Claude retrieves real-time information from the web), and Dify Workflows & Agents (Claude runs multi-step automations built on the open-source Dify platform, hosted inside your VPC). Because Dify supports MCP and standard APIs, Claude can be connected to virtually any business system through Bridge, including Salesforce, Jira, Zendesk, Confluence, SAP, or custom internal tools, and use them inside workflows or agents.

No. Every MCP plugin runs inside your AWS VPC. When Claude queries SharePoint, runs a SQL query, or executes a Dify workflow, the data flows between your systems and Claude on Bedrock, all inside your AWS perimeter. Web Search & Scrape is the only plugin that reaches external websites, and only for the specific URLs requested by users. Nothing is sent to Moterra, Anthropic, or third-party services beyond what the plugin explicitly needs.

By default, every user prompt, model response, plugin call, and administrative action is captured in AWS CloudTrail plus a dedicated log bucket in your AWS account, with retention and access controlled by you. For enterprise deployments, Bridge can forward logs directly to your SIEM (Splunk, Sentinel, Datadog, or others) or to a custom log destination. The logs live in your environment, so you can meet DORA, GDPR, and internal audit requirements without depending on Moterra.

Both. Moterra builds the initial set of agents for standard use cases such as document analysis, meeting summaries, and structured data extraction during onboarding. After go-live, your team gets proper training from Moterra and can build additional agents through Dify’s visual workflow builder: simple chatbots and Q&A agents require no coding, while more complex workflows benefit from basic technical understanding of APIs and conditional logic. For anything advanced, Priority and Enterprise Support tiers include consulting hours for custom agent development.

One to two weeks alongside Private Claude Cowork, when deployed together. Bridge shares the same AWS account, VPC, and identity integration as Cowork, so most of the setup work is done once. If you already have Cowork running and are adding Bridge on top, expect about one week for admin console setup, plugin configuration, and initial agent build-out.