Skip to content
Plur-e

Plur-e MCP · Security & governance

Permissions, audit and data handling for AI agents on your ERP

An agent is only as safe as the boundaries around it. These are ours.

Quick answer

Plur-e MCP servers authenticate each user with Microsoft Entra ID and call Business Central and Dataverse with that user's token, so existing permissions apply; tools are read-only by default and write tools are enabled per role with optional human approval; every request and tool call is logged with user, inputs and outputs; prompts and results are processed by the Claude API under Anthropic's commercial terms and are not used to train models; servers run in Plur-e's Azure or the customer's Azure tenant.

Identity and permissions

Identity and permissions
AuthenticationMicrosoft Entra ID, per user (OAuth 2.0, on-behalf-of flow). VALIDAR exact flow
AuthorizationBusiness Central permission sets and Dataverse security roles apply to every call
Service accountsOnly for scheduled/background tools, with least-privilege permission sets

Tool scopes and approvals

Tool scopes and approvals
DefaultRead-only tools
Write toolsEnabled per customer and role; annotated as non-read-only in the tool schema
ApprovalsOptional human approval before posting documents; approver and payload logged
LimitsRate limits per user and tenant; maximum rows per response

Audit and monitoring

Audit and monitoring
LogsRequest, tool name, inputs, outputs, user, timestamp, latency
RetentionVALIDAR: default retention period and export options
AlertsFailed authentications, denied writes, unusual volumes

Data handling

Data handling
Model providerClaude API (Anthropic). Data submitted through the API is not used to train models under Anthropic's commercial terms
ResidencyBusiness Central and Dataverse data stays in your Microsoft tenant; the MCP server runs in Azure (Plur-e or yours)
SubprocessorsMicrosoft Azure, Anthropic. VALIDAR complete list for the DPA
DeletionLogs deleted on request; no customer data stored beyond logs and cache. VALIDAR cache policy

FAQ

Frequently asked questions

Start with a two-hour discovery workshop

Bring the questions your team asks every day. We map them to tools, permissions and a pilot plan.