mongodb-atlas-stream-processing

What it does

Manages MongoDB Atlas Stream Processing (ASP) workflows, handling workspace provisioning, data source/sink connections, processor lifecycle operations, debugging diagnostics, and tier sizing. Supports Kafka, Atlas clusters, S3, HTTPS, and Lambda integrations for streaming data workloads and event processing. NOT for general MongoDB queries or Atlas cluster management. Requires MongoDB MCP Server with Atlas API credentials.

How it works

The skill provides four MCP tools to operate ASP pipelines:

• atlas-streams-discover: ALL read operations (list-workspaces, inspect-workspace, list-connections, inspect-connection, list-processors, inspect-processor, diagnose-processor, get-networking) with pagination and response formats.
• atlas-streams-build: ALL create operations for resources: workspace, connection, processor, privatelink. Field mapping rules specify which fields to fill per resource type.
• atlas-streams-manage: ALL update/state operations (start-processor, stop-processor, modify-processor, update-workspace, update-connection, accept-peering, reject-peering) with state pre-checks and processor lifecycle rules.
• atlas-streams-teardown: ALL delete operations with safety behavior (auto-stops processors, blocks deletions if referenced, cascading workspace deletes) and a pre-delete workspace confirmation workflow.

Critical validation step:

• You MUST call search-knowledge before composing any processor pipeline. This includes field validation queries and example pattern references, plus consulting https://github.com/mongodb/ASP_example for non-trivial processors. Examples and quickstarts are listed (e.g., 00_hello_world.json, 01_changestream_basic.json).

Pipeline rules:

• Several constraints and required fields per stage are documented (e.g., sources, sinks, and enrichment patterns; specific field requirements for $source, $emit, $https, $externalFunction, $validate, $lookup, etc.).
• AWS/Azure/GCP region mappings and Atlas region naming conventions are included to avoid region format errors.

MCP Tool Behaviors:

• Elicitation: missing sensitive fields are auto-collected by the build tool during connection creation.
• Auto-normalization: specific fields auto-convert (e.g., bootstrapServers, schemaRegistryUrls).
• Workspace defaults and region mappings are outlined for correct configuration.

Safety and teardown rules are specified for processor, connection, and workspace deletions, including pre-deletion inspection steps for workspaces.

When to use it

Use when your workflow involves Atlas Stream Processing tasks and you have the MongoDB MCP Server connected with Atlas API credentials. Use the discovery tool to inventory, the build tool to create resources, the manage tool to start/stop/modify, and the teardown tool to delete, following the safety workflow for workspaces.

What it can touch

The skill interacts with Atlas Stream Processing resources via MCP tools: atlas-streams-discover, atlas-streams-build, atlas-streams-manage, atlas-streams-teardown. These tools operate on workspaces, connections, processors, and privatelink/peering resources, requiring fields as specified in the tool matrices.

Caveats

Requires MongoDB MCP Server connected with Atlas API credentials (apiClientId and apiClientSecret). All operations require an Atlas project ID. PrivateLink changes are project-level; networking config cannot be modified after creation without deletion and recreation. Before deleting a workspace, explicit confirmation is required.