Last9 MCP Server

Last9 MCP Server

What is Last9 MCP Server about?

Enables AI agents to retrieve live production observability data (logs, metrics, traces, alerts, and drop‑rule configurations) from Last9, allowing them to understand the current state of services and suggest or apply fixes directly from development environments.

How to use Last9 MCP Server?

1. Install – preferred method is via npm:

   npx @last9/mcp-server
   

   (or install globally with npm install -g @last9/mcp-server or via Homebrew brew install last9-mcp).
2. Configure environment variables:

   export LAST9_BASE_URL=<your_last9_otlp_host>
   export LAST9_AUTH_TOKEN=<your_last9_otlp_auth_token>
   export LAST9_REFRESH_TOKEN=<your_last9_write_refresh_token>
   

3. Run the server (the command above starts the MCP server in stdio mode).
4. Integrate with IDEs – add the following configuration to Claude Desktop, Cursor, Windsurf, or VS Code:

   {
     "mcpServers": {
       "last9": {
         "command": "/opt/homebrew/bin/last9-mcp",
         "env": {
           "LAST9_BASE_URL": "<last9_otlp_host>",
           "LAST9_AUTH_TOKEN": "<last9_otlp_auth_token>",
           "LAST9_REFRESH_TOKEN": "<last9_write_refresh_token>"
         }
       }
     }
   }
   

   Adjust the command path if you use the npx launch method.

Key features

• Observability & APM tools: get exceptions, service summaries, environment lists, performance details, operations summary, dependency graph.
• Prometheus/PromQL tools: range queries, instant queries, label discovery.
• Logs management: fetch logs, service‑specific logs, view and add drop‑rules.
• Alert management: retrieve alert configurations and currently firing alerts.
• Cross‑IDE support: works with Claude desktop, Cursor, Windsurf, and VS Code (preview).

Use cases

• Automated debugging: AI agents can fetch the exact logs and metrics surrounding a failure and propose code changes.
• Performance monitoring: Retrieve real‑time throughput, error rates, and latency to guide optimization suggestions.
• Alert‑driven remediation: When an alert fires, the model can automatically query related traces and logs to generate a fix.
• Developer workflow enhancement: Seamlessly integrate production context into IDE chat panels, reducing context‑switching.

FAQ

Q: Do I need a Last9 account? A: Yes, an active Last9 workspace with OTel integration and API tokens is required.

Q: Which IDEs are supported? A: Claude Desktop, Cursor, Windsurf, and VS Code (from version 1.99, preview).

Q: Can I run the server locally without installation? A: Yes, using npx @last9/mcp-server launches the server directly without a global install.

Q: How are credentials secured? A: Provide them via environment variables; the server does not store them on disk.

Q: What limits apply to queries? A: Each tool defines default limits (e.g., 20 logs, 60‑minute look‑back) which can be overridden via parameters.

Last9 MCP Server

A Model Context Protocol server implementation for Last9 that enables AI agents to seamlessly bring real-time production context — logs, metrics, and traces — into your local environment to auto-fix code faster.

• View demo
• Read our announcement blog post

Status

Works with Claude desktop app, or Cursor, Windsurf, and VSCode (Github Copilot) IDEs. Implements the following MCP tools:

Observability & APM Tools:

• get_exceptions: Get the list of exceptions.
• get_service_summary: Get service summary with throughput, error rate, and response time.
• get_service_environments: Get available environments for services.
• get_service_performance_details: Get detailed performance metrics for a service.
• get_service_operations_summary: Get operations summary for a service.
• get_service_dependency_graph: Get service dependency graph showing incoming/outgoing dependencies.

Prometheus/PromQL Tools:

• promptheus_range_query: Execute PromQL range queries for metrics data.
• prometheus_instant_query: Execute PromQL instant queries for metrics data.
• prometheus_label_values: Get label values for PromQL queries.
• prometheus_labels: Get available labels for PromQL queries.

Logs Management:

• get_logs: Get logs filtered by service name and/or severity level.
• get_drop_rules: Get drop rules for logs that determine what logs get filtered out at Last9 Control Plane
• add_drop_rule: Create a drop rule for logs at Last9 Control Plane
• get_service_logs: Get raw log entries for a specific service over a time range. Can apply filters on severity and body.

Alert Management:

• get_alert_config: Get alert configurations (alert rules) from Last9.
• get_alerts: Get currently active alerts from Last9 monitoring system.

Alert Management:

• get_alert_config: Get alert configurations (alert rules) from Last9.
• get_alerts: Get currently active alerts from Last9 monitoring system.

Tools Documentation

get_exceptions

Retrieves server-side exceptions over a specified time range.

Parameters:

• limit (integer, optional): Maximum number of exceptions to return. Default: 20.
• lookback_minutes (integer, recommended): Number of minutes to look back from now. Default: 60. Examples: 60, 30, 15.
• start_time_iso (string, optional): Start time in ISO format (YYYY-MM-DD HH:MM:SS). Leave empty to use lookback_minutes.
• end_time_iso (string, optional): End time in ISO format (YYYY-MM-DD HH:MM:SS). Leave empty to default to current time.
• span_name (string, optional): Name of the span to filter by.

get_service_summary

Get service summary over a given time range. Includes service name, environment, throughput, error rate, and response time. All values are p95 quantiles over the time range.

Parameters:

• start_time_iso (string, optional): Start time in ISO format (YYYY-MM-DD HH:MM:SS). Leave empty to default to end_time_iso - 1 hour.
• end_time_iso (string, optional): End time in ISO format (YYYY-MM-DD HH:MM:SS). Leave empty to default to current time.
• env (string, optional): Environment to filter by. Defaults to 'prod'.

get_service_environments

Get available environments for services. Returns an array of environments that can be used with other APM tools.

Parameters:

• start_time_iso (string, optional): Start time in ISO format (YYYY-MM-DD HH:MM:SS). Leave empty to default to end_time_iso - 1 hour.
• end_time_iso (string, optional): End time in ISO format (YYYY-MM-DD HH:MM:SS). Leave empty to default to current time.

Note: All other APM tools that retrieve service information (like get_service_performance_details, get_service_dependency_graph, get_service_operations_summary, get_service_summary) require an env parameter. This parameter must be one of the environments returned by this tool. If this tool returns an empty array, use an empty string "" for the env parameter.

get_service_performance_details

Get detailed performance metrics for a specific service over a given time range.

Parameters:

• service_name (string, required): Name of the service to get performance details for.
• start_time_iso (string, optional): Start time in ISO format (YYYY-MM-DD HH:MM:SS). Leave empty to default to now - 60 minutes.
• end_time_iso (string, optional): End time in ISO format (YYYY-MM-DD HH:MM:SS). Leave empty to default to current time.
• env (string, optional): Environment to filter by. Defaults to 'prod'.

get_service_operations_summary

Get a summary of operations inside a service over a given time range. Returns operations like HTTP endpoints, database queries, messaging producer and HTTP client calls.

Parameters:

• service_name (string, required): Name of the service to get operations summary for.
• start_time_iso (string, optional): Start time in ISO format (YYYY-MM-DD HH:MM:SS). Leave empty to default to now - 60 minutes.
• end_time_iso (string, optional): End time in ISO format (YYYY-MM-DD HH:MM:SS). Leave empty to default to current time.
• env (string, optional): Environment to filter by. Defaults to 'prod'.

get_service_dependency_graph

Get details of the throughput, response times and error rates of incoming, outgoing and infrastructure components of a service. Useful for analyzing cascading effects of errors and performance issues.

Parameters:

• service_name (string, optional): Name of the service to get the dependency graph for.
• start_time_iso (string, optional): Start time in ISO format (YYYY-MM-DD HH:MM:SS). Leave empty to default to now - 60 minutes.
• end_time_iso (string, optional): End time in ISO format (YYYY-MM-DD HH:MM:SS). Leave empty to default to current time.
• env (string, optional): Environment to filter by. Defaults to 'prod'.

promptheus_range_query

Perform a Prometheus range query to get metrics data over a specified time range. Recommended to check available labels first using prometheus_labels tool.

Parameters:

• query (string, required): The range query to execute.
• start_time_iso (string, optional): Start time in ISO format (YYYY-MM-DD HH:MM:SS). Leave empty to default to now - 60 minutes.
• end_time_iso (string, optional): End time in ISO format (YYYY-MM-DD HH:MM:SS). Leave empty to default to current time.

prometheus_instant_query

Perform a Prometheus instant query to get metrics data at a specific point in time. Typically should use rollup functions like sum_over_time, avg_over_time, quantile_over_time over a time window.

Parameters:

• query (string, required): The instant query to execute.
• time_iso (string, optional): Time in ISO format (YYYY-MM-DD HH:MM:SS). Leave empty to default to current time.

prometheus_label_values

Return the label values for a particular label and PromQL filter query. Similar to Prometheus /label_values call.

Parameters:

• match_query (string, required): A valid PromQL filter query.
• label (string, required): The label to get values for.
• start_time_iso (string, optional): Start time in ISO format (YYYY-MM-DD HH:MM:SS). Leave empty to default to now - 60 minutes.
• end_time_iso (string, optional): End time in ISO format (YYYY-MM-DD HH:MM:SS). Leave empty to default to current time.

prometheus_labels

Return the labels for a given PromQL match query. Similar to Prometheus /labels call.

Parameters:

• match_query (string, required): A valid PromQL filter query.
• start_time_iso (string, optional): Start time in ISO format (YYYY-MM-DD HH:MM:SS). Leave empty to default to now - 60 minutes.
• end_time_iso (string, optional): End time in ISO format (YYYY-MM-DD HH:MM:SS). Leave empty to default to current time.

get_logs

Gets logs filtered by optional service name and/or severity level within a specified time range.

Parameters:

• service (string, optional): Name of the service to get logs for.
• severity (string, optional): Severity of the logs to get.
• lookback_minutes (integer, recommended): Number of minutes to look back from now. Default: 60. Examples: 60, 30, 15.
• start_time_iso (string, optional): Start time in ISO format (YYYY-MM-DD HH:MM:SS). Leave empty to use lookback_minutes.
• end_time_iso (string, optional): End time in ISO format (YYYY-MM-DD HH:MM:SS). Leave empty to default to current time.
• limit (integer, optional): Maximum number of logs to return. Default: 20.

get_drop_rules

Gets drop rules for logs, which determine what logs get filtered out from reaching Last9.

add_drop_rule

Adds a new drop rule to filter out specific logs at Last9 Control Plane

Parameters:

• name (string, required): Name of the drop rule.
• filters (array, required): List of filter conditions to apply. Each filter has:
  • key (string, required): The key to filter on. Only attributes and resource.attributes keys are supported. For resource attributes, use format: resource.attributes[key_name] and for log attributes, use format: attributes[key_name] Double quotes in key names must be escaped.
  • value (string, required): The value to filter against.
  • operator (string, required): The operator used for filtering. Valid values:
    • "equals"
    • "not_equals"
  • conjunction (string, required): The logical conjunction between filters. Valid values:
    • "and"

get_alert_config

Get alert configurations (alert rules) from Last9. Returns all configured alert rules including their conditions, labels, and annotations.

Parameters:

None - This tool retrieves all available alert configurations.

Returns information about:

• Alert rule ID and name
• Primary indicator being monitored
• Current state and severity
• Algorithm used for alerting
• Entity ID and organization details
• Properties and configuration
• Creation and update timestamps
• Group timeseries notification settings

get_alerts

Get currently active alerts from Last9 monitoring system. Returns all alerts that are currently firing or have fired recently within the specified time window.

Parameters:

• timestamp (integer, optional): Unix timestamp for the query time. Leave empty to default to current time.
• window (integer, optional): Time window in seconds to look back for alerts. Defaults to 900 seconds (15 minutes). Range: 60-86400 seconds.

Returns information about:

• Alert rule details (ID, name, group, type)
• Current state and severity
• Last fired timestamp and duration
• Rule properties and configuration
• Alert instances with current values
• Metric degradation information
• Group labels and annotations for each instance

get_service_logs

Get raw log entries for a specific service over a time range. Can apply filters on severity and body.

Parameters:

• service (string, required): Name of the service to get logs for.
• start_time_iso (string, optional): Start time in ISO format (YYYY-MM-DD HH:MM:SS). Leave empty to default to now - lookback_minutes.
• end_time_iso (string, optional): End time in ISO format (YYYY-MM-DD HH:MM:SS). Leave empty to default to current time.
• lookback_minutes (integer, recommended): Number of minutes to look back from now. Default: 60. Examples: 60, 30, 15.
• limit (integer, optional): Maximum number of logs to return. Default: 20.
• severity_filters (array, optional): List of severity filters to apply. Valid values: "debug", "info", "warn", "error", "fatal".
• body_filters (array, optional): List of body filters to apply.

Installation

You can install the Last9 Observability MCP server using either:

Homebrew

# Add the Last9 tap
brew tap last9/tap

# Install the Last9 MCP CLI
brew install last9-mcp

NPM

# Install globally
npm install -g @last9/mcp-server

# Or run directly with npx
npx @last9/mcp-server

Configuration

Environment Variables

The Last9 MCP server requires the following environment variables:

• LAST9_BASE_URL: (required) Last9 API URL from OTel integration
• LAST9_AUTH_TOKEN: (required) Authentication token for Last9 MCP server from OTel integration
• LAST9_REFRESH_TOKEN: (required) Refresh Token with Write permissions, needed for accessing control plane APIs from API Access

Usage with Claude Desktop

Configure the Claude app to use the MCP server:

1. Open the Claude Desktop app, go to Settings, then Developer
2. Click Edit Config
3. Open the claude_desktop_config.json file
4. Copy and paste the server config to your existing file, then save
5. Restart Claude

{
  "mcpServers": {
    "last9": {
      "command": "/opt/homebrew/bin/last9-mcp",
      "env": {
        "LAST9_BASE_URL": "<last9_otlp_host>",
        "LAST9_AUTH_TOKEN": "<last9_otlp_auth_token>",
        "LAST9_REFRESH_TOKEN": "<last9_write_refresh_token>"
      }
    }
  }
}

Usage with Cursor

Configure Cursor to use the MCP server:

1. Open Cursor, go to Settings, then Cursor Settings
2. Select MCP on the left
3. Click Add "New Global MCP Server" at the top right
4. Copy and paste the server config to your existing file, then save
5. Restart Cursor

{
  "mcpServers": {
    "last9": {
      "command": "/opt/homebrew/bin/last9-mcp",
      "env": {
        "LAST9_BASE_URL": "<last9_otlp_host>",
        "LAST9_AUTH_TOKEN": "<last9_otlp_auth_token>",
        "LAST9_REFRESH_TOKEN": "<last9_write_refresh_token>"
      }
    }
  }
}

Usage with Windsurf

Configure Windsurf to use the MCP server:

1. Open Windsurf, go to Settings, then Developer
2. Click Edit Config
3. Open the windsurf_config.json file
4. Copy and paste the server config to your existing file, then save
5. Restart Windsurf

{
  "mcpServers": {
    "last9": {
      "command": "/opt/homebrew/bin/last9-mcp",
      "env": {
        "LAST9_BASE_URL": "<last9_otlp_host>",
        "LAST9_AUTH_TOKEN": "<last9_otlp_auth_token>",
        "LAST9_REFRESH_TOKEN": "<last9_write_refresh_token>"
      }
    }
  }
}

Usage with VS Code

Note: MCP support in VS Code is available starting v1.99 and is currently in preview. For advanced configuration options and alternative setup methods, view the VS Code MCP documentation.

1. Open VS Code, go to Settings, select the User tab, then Features, then Chat
2. Click "Edit settings.json"
3. Copy and paste the server config to your existing file, then save
4. Restart VS Code

{
  "mcp": {
    "servers": {
      "last9": {
        "type": "stdio",
        "command": "/opt/homebrew/bin/last9-mcp",
        "env": {
          "LAST9_BASE_URL": "<last9_otlp_host>",
          "LAST9_AUTH_TOKEN": "<last9_otlp_auth_token>",
          "LAST9_REFRESH_TOKEN": "<last9_write_refresh_token>"
        }
      }
    }
  }
}

Badges