Skip to main content
The Whawit MCP (Model Context Protocol) server enables AI assistants to access your observability data, search project knowledge, and analyze incidents programmatically.

​
Overview

MCP is an open protocol that allows AI assistants to interact with external tools and data sources. The Whawit MCP server exposes eight tools that give AI assistants full access to:
  • Project knowledge base search
  • DevOps analysis and log querying
  • Incident details and correlated analyses
  • AI-optimized prompts for fixing issues

​
Hosted Server

Whawit provides a hosted MCP server that requires no local installation:
Endpoint
https://mcp.whawit.ai/sse
The hosted server is recommended for most users. It stays up-to-date automatically and requires no maintenance.

​
Quick Setup

Add to ~/.cursor/mcp.json:
{
  "mcpServers": {
    "WHAWIT": {
      "url": "https://mcp.whawit.ai/sse",
      "headers": {
        "X-API-Key": "YOUR_WHAWIT_API_KEY"
      }
    }
  }
}

​
Getting Your API Key

Generate your API key from the Whawit web app:
  1. Go to app.whawit.ai
  2. Navigate to Settings > User Settings
  3. Generate a new API key
Keep your API key secure. Never commit it to version control or share it publicly.

​
Available Tools

The MCP server exposes eight tools for AI assistants: Search the project knowledge base for relevant context about architecture, business logic, and implementation details.
​
query
string
required
The search query to find relevant context.
​
project
string
required
The project URN to search in (e.g., urn:project:abc123).
​
types
array
Types of contexts to search: repo, project, user, other. Defaults to all.
​
maxResults
number
default:"3"
Maximum number of results to return. Range: 1-5.
Example prompt:
Search for how authentication works in project urn:project:abc123

​
whawit_devops_analysis

Analyze logs, metrics, and system health from connected observability providers.
​
query
string
required
What to analyze (e.g., “redis errors”, “API failures”, “high latency”).
​
project
string
required
The project URN to analyze.
​
from
string
default:"1h"
Time range start (e.g., 15m, 1h, 24h).
​
to
string
default:"now"
Time range end.
​
verbosity
string
default:"concise"
Output verbosity: concise or detailed.
Example prompt:
Check for errors in the last 15 minutes in project urn:project:abc123

​
whawit_project_info

Get information about a project including its integrations, contexts, and configuration.
​
project
string
required
The project URN to get info for.

​
whawit_list_projects

List all available projects for the authenticated user. This tool takes no parameters.

​
whawit_get_analysis_full

Get complete analysis data with full JSON payload and formatted markdown.
​
historyUrn
string
required
The URN of the analysis to fetch (e.g., urn:history:abc123).
​
format
string
default:"both"
Output format: markdown, json, or both.
​
includeRawLogs
boolean
default:"false"
Include raw log events in output.
​
maxLength
number
default:"15000"
Maximum output length in characters.

​
whawit_get_incident_full

Get full incident details with optional correlated analysis.
​
project
string
required
The project URN to fetch incidents from.
​
incidentUrn
string
Specific incident URN to fetch.
​
severity
string
Filter by severity: critical, high, medium, low. Gets first matching open incident.
​
includeCorrelatedAnalysis
boolean
default:"true"
Include the correlated analysis if available.

​
whawit_generate_fix_prompt

Generate an AI-ready prompt for fixing issues identified in an analysis.
​
historyUrn
string
required
The URN of the analysis to generate a fix prompt for.
​
focus
string
default:"all"
Focus area: code, infrastructure, or all.

​
whawit_codebase_analysis_prompt

Generate a prompt for analyzing a codebase based on project context.
​
project
string
required
The project URN to analyze.
​
query
string
Optional specific query to focus the analysis on.
​
includeArchitecture
boolean
default:"true"
Include architecture/overview contexts.

​
MCP Resources

The server also exposes projects as MCP resources: 
whawit://project/{urn}
Reading a project resource returns:
  • Project details (name, description, URN)
  • Connected providers
  • Available contexts with titles and types

​
Authentication

The MCP server uses API key authentication via the X-API-Key header: 
X-API-Key: YOUR_WHAWIT_API_KEY
Generate your API key in the Whawit web app under Settings > User Settings.

​
Transport Protocols

The Whawit MCP server supports two transport protocols:
ProtocolEndpointUse Case
SSE (Server-Sent Events)/sseTraditional MCP clients
StreamableHTTP/mcpModern clients like Cursor
Both endpoints support the full MCP protocol including tools and resources.

​
Example Workflows

Whawit MCP tools in action

Using Whawit MCP tools in Cursor to analyze incidents

​
Investigating an Incident

  1. Ask Claude to list your projects:
    “List my Whawit projects”
  2. Get incident details:
    “Get the latest critical incident from project urn:project:abc123”
  3. Generate a fix prompt:
    “Generate a fix prompt for the incident analysis”
  4. Let Claude implement the fix based on the context

​
Searching Documentation

“Search the project knowledge base for how the payment processing flow works in urn:project:payments”

​
Running DevOps Analysis

“Analyze Redis connection errors in the last 30 minutes for urn:project:backend”

​
Troubleshooting

  1. Verify your MCP configuration file syntax is valid JSON
  2. Restart Claude Desktop or Cursor after configuration changes
  3. Check that the access token is valid and not expired
  4. Ensure the MCP server URL is reachable from your network
  1. Verify your API key is correct
  2. Ensure the X-API-Key header is set correctly
  3. Generate a new API key from Settings > User Settings if needed
  1. Verify the project URN format is correct (urn:project:...)
  2. Ensure you have access to the project in Whawit
  3. Run whawit_list_projects to see available projects

​
Next Steps

VS Code Extension

Install the Whawit extension for integrated observability in your editor.

GitHub Copilot

Use @whawit in GitHub Copilot Chat for AI-assisted analysis.