KLogic
🔌 API Access

Kafka REST API & Programmatic Access

Every KLogic feature is accessible via a fully documented REST API. Automate operations, build integrations, embed Kafka metrics in your own tooling, and connect AI assistants directly to your cluster state via MCP.

Kafka Visibility Shouldn't Stop at the UI

Without API access, Kafka data stays locked in dashboards and can't reach the systems that need it

Manual Operations

Teams spend hours manually checking Kafka health in dashboards instead of automating checks in their CI/CD pipelines and runbooks.

Isolated Tooling

Kafka metrics live in one tool, infrastructure metrics in another, and application metrics in a third — no single system can correlate across them.

AI Tools Without Context

AI coding assistants can't help with Kafka issues because they have no access to live cluster state — engineers have to copy-paste data manually.

Full Programmatic Control

Build, automate, and integrate with the same API that powers the KLogic dashboard

Complete REST API

Full Feature Parity

Every action available in the KLogic UI is exposed via REST — metrics, alerts, forecasts, topology, and configuration

OpenAPI 3.0 Specification

Complete spec at /api/openapi.json — import into Postman, generate clients in TypeScript, Python, or Go

Consistent JSON Responses

All endpoints return structured JSON with standardized pagination, error codes, and metadata envelopes

GET /api/v1/clusters/:id/consumer-groups
{
"data": [
{
"group_id": "payment-processor",
"state": "Stable",
"total_lag": 1423,
"members": 8
}
],
"meta": { "total": 42, "page": 1 }
}
API Key Management

kl_prod_7f3a...c29e

Production · Read-only · Expires in 89 days

Active

kl_ci_2b9d...f14a

CI/CD · Read-write · No expiry

Active

kl_mcp_9e1c...a73b

MCP Integration · Read-only · Expires in 30 days

Expiring

API Key Management

Scoped Permissions

Issue read-only, read-write, or admin keys per workspace — least-privilege access for every integration

Configurable Expiration

Set expiry from 1 day to never, with dashboard warnings and optional auto-notify before a key expires

Zero-Downtime Rotation

Generate a replacement key and let both coexist during a grace period — rotate without service interruption

MCP Integration for AI Tools

Connect Claude, Cursor & More

Configure KLogic as an MCP server in your AI assistant — query live Kafka state directly from your chat interface

Natural Language Kafka Queries

Ask "which consumer groups are lagging?" or "is broker-3 healthy?" and get answers backed by live cluster data

Read-Only Safety

MCP connections use read-only API keys by default — AI tools can observe but never modify your cluster configuration

MCP Configuration (claude_desktop_config.json)

{
"mcpServers": {
"klogic": {
"command": "npx",
"args": ["klogic-mcp"],
"env": {
"KLOGIC_API_KEY": "kl_mcp_...",
"KLOGIC_URL": "https://app.klogic.io"
}
}
}
}

Frequently Asked Questions

The KLogic REST API exposes every feature available in the UI: cluster metadata, broker health, topic metrics, consumer group lag, schema registry data, Kafka Connect status, alert rules, watchdog events, and forecasts. You can query, create, update, and delete resources programmatically with full parity to the web interface.

API keys are scoped to a workspace and can be further restricted by permission level: read-only, read-write, or admin. Read-only keys are ideal for dashboards and monitoring integrations. Read-write keys support automation and CI/CD pipelines. Admin keys are required for configuration changes and user management.

In the KLogic dashboard under Settings > API Keys, you can generate a replacement key for any existing key. The new key is issued immediately and the old key remains valid for a configurable grace period (default 24 hours) to allow a zero-downtime rotation. After the grace period, the old key is automatically revoked.

MCP (Model Context Protocol) is an open standard that allows AI assistants like Claude, Cursor, and other AI tools to connect to KLogic as a context source. Engineers who use AI coding assistants can configure KLogic as an MCP server, enabling their AI tools to query live Kafka metrics and answer questions about cluster state directly in their development environment.

Yes. KLogic publishes a complete OpenAPI 3.0 specification at /api/openapi.json. You can import it into Postman, Insomnia, or any OpenAPI-compatible client. Client SDKs for TypeScript, Python, and Go can be generated from the spec using standard tooling.

Yes. Common CI/CD use cases include querying consumer lag before and after deployments to verify processing health, checking topic configuration matches expected values, creating or updating alert rules as part of infrastructure-as-code workflows, and retrieving cluster topology for automated documentation generation.

Build On Top of Your Kafka Data

The KLogic API gives you programmatic access to every metric, event, and insight in the platform. Automate operations, integrate with your existing tools, and bring AI assistance to your Kafka workflows.

Free 14-day trial • No credit card required • Setup in 5 minutes