mcpField
C
Coralogix

Coralogix

Remote
coralogix.comData & Analytics1Updated 4d ago

About Coralogix

Coralogix MCP Server is a hosted connection to Coralogix observability data for teams investigating application and infrastructure behavior. MCP clients can query telemetry and manage selected observability objects; the public repository contains documentation and connection metadata, not the server implementation.

Key features

  1. 01Query logs, spans, and datasets
  2. 02Run PromQL metric queries
  3. 03Investigate alerts and cases
  4. 04Explore service maps and catalogs
  5. 05Search dashboards and saved views
  6. 06Inspect RUM and AI Center data

Use cases

  1. 01Find frequent errors in a service
  2. 02Trace an alert through its case timeline
  3. 03Check latency percentiles over a time range

MCP support

The remote MCP server exposes Coralogix observability queries and management tools to MCP clients, including logs, metrics, traces, RUM, alerts, cases, dashboards, and service catalog data. It is hosted by Coralogix and supports OAuth 2.1 or bearer API key authentication.

Configuration

Remote servers need no local install. Pick your client, then copy the config or install in one click.

Endpointhttps://api.{region}.coralogix.com/mgmt/api/v1/mcp
Copy install command
Claude CodeTerminal
Cursor.cursor/mcp.json
VS Code.vscode/mcp.json
Claude Desktopclaude_desktop_config.json
Windsurf~/.codeium/windsurf/mcp_config.json

Run commands in a terminal; paste JSON into the client’s MCP config and restart. Servers that need sign-in will prompt on first connect.

All endpoints
Streamable HTTPhttps://api.{region}.coralogix.com/mgmt/api/v1/mcpAuth required

Tools

Details

From the project README

Coralogix MCP Server

Query and investigate your Coralogix observability data — logs, spans, metrics, RUM, alerts, cases, dashboards and service catalog — from any Model Context Protocol client.

The server is fully hosted and remote. There is nothing to install, build, or run.

https://api.<region>.coralogix.com/mgmt/api/v1/mcp

About this repository. This is the public home for Coralogix MCP documentation, connection examples, and registry metadata. The server implementation is maintained privately by Coralogix; this repo carries no server source code. See Contributing.


Quick start

Pick your client, substitute your region, and connect. OAuth runs in the browser on first use — no API key needed.

Claude Code

claude mcp add --transport http coralogix https://api.eu2.coralogix.com/mgmt/api/v1/mcp

Cursor — .cursor/mcp.json

{
  "mcpServers": {
    "coralogix": {
      "url": "https://api.eu2.coralogix.com/mgmt/api/v1/mcp"
    }
  }
}

Codex

codex mcp add coralogix --url "https://api.eu2.coralogix.com/mgmt/api/v1/mcp"

Full, copy-pasteable configurations for these and other clients are in examples/clients/.

Then ask your assistant:

"Show me the most frequent errors for service payments-api in the last hour." "Why did this alert fire? Walk me through the case timeline." "What's the p99 latency for checkout over the past 24 hours?"

More in examples/prompts/.


What it can do

71 tools across 21 capability domains — 65 by default, plus the six Olly agent tools when x-allow-costly-use is set — and 20 skills that teach the model Coralogix query languages and investigation workflows on demand.

DomainWhat it covers
Logs, Spans & DatasetsQuery logs, spans and named datasets with DataPrime
MetricsPromQL instant and range queries, metric and label discovery
Alerts & CasesAlert definitions, alert events, case triage and lifecycle
Service CatalogRED metrics — error rate, latency percentiles, throughput, saturation
Service MapService-to-service dependency graph
InfrastructureResource inventory, health history, raw resource data
RUMReal user monitoring — Web Vitals, frontend errors, sessions
DashboardsList, search, create, replace and delete dashboards and folders
Saved ViewsManage saved log views and view folders
Data UsageBillable ingestion volumes, quota attribution, usage trends
AI CenterObserve and evaluate GenAI applications, guardrails, model pricing
Field DiscoverySemantic search across log and span fields
SchemasDataset schemas and available fields
DataPrime LanguageCommand and function reference, construct guidance
DocumentationSearch and fetch Coralogix docs
UtilitiesCurrent UTC time, for grounding relative time ranges
Olly AgentCoralogix's natural-language observability agent (opt-in)

Tools declare MCP annotations (readOnlyHint, destructiveHint, openWorldHint) so clients know what needs approval. Most are read-only; the manage_* tools that mutate state are flagged destructive.


Documentation

Getting startedConnect, verify, and run a first query
RegionsEndpoint URL for every Coralogix region
AuthenticationOAuth 2.1 and API key, permissions, tenancy
Tool catalogAll 71 current tools, and the 12 deprecated ones, with annotations
SkillsThe 20 progressive-disclosure skills
Headers and versioningmcp-version, x-allow-costly-use
TroubleshootingCommon failures and what they mean
Registry publishingHow server.json is published and verified

Agents: start at llms.txt.


Registry

Registry metadata lives in server.json, following the official MCP registry schema.

Registry namecom.coralogix/mcp-server
Transportstreamable-http
Server version2.0.0
AuthenticationOAuth 2.1 (OIDC) or bearer API key

Support

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.