Conversational Api Debugger

Debug REST API failures using OpenAPI specs and HTTP logs

An MCP server that helps developers quickly identify and fix API issues by analyzing OpenAPI specifications, ingesting HTTP logs, explaining failures, and generating reproducible test commands.

What It Does

This plugin transforms API debugging from guesswork into systematic analysis:

1. Load OpenAPI Specs - Parse API documentation (JSON/YAML)
2. Ingest HTTP Logs - Import request/response data (HAR, JSON)
3. Explain Failures - Analyze why API calls failed with root cause analysis
4. Generate Repros - Create cURL/HTTPie/fetch commands to reproduce issues

Installation

/plugin install conversational-api-debugger@claude-code-plugins-plus

Features

4 Powerful MCP Tools

1. load_openapi

Load and parse OpenAPI 3.x specifications.

{
  "filePath": "/path/to/openapi.yaml",
  "name": "my-api"
}

Returns:

• API title, version, description
• Base URL and servers
• Complete endpoint list
• Endpoint count and structure

2. ingest_logs

Import HTTP request/response logs for analysis.

{
  "filePath": "/path/to/requests.har",
  "format": "har"
}

Or provide logs directly:

{
  "logs": [
    {
      "timestamp": "2025-10-10T12:00:00Z",
      "method": "POST",
      "url": "https://api.example.com/users",
      "statusCode": 400,
      "requestBody": { "name": "John" },
      "responseBody": { "error": "Missing required field: email" }
    }
  ]
}

Returns:

• Total requests ingested
• Success/failure breakdown
• Status code distribution
• Method distribution
• Top errors (first 10)

3. explain_failure

Analyze why an API call failed.

{
  "logIndex": 0,
  "specName": "my-api"
}

Returns:

• Severity: critical | high | medium | low
• Possible Causes: List of likely root causes
• Suggested Fixes: Actionable remediation steps
• Matching Endpoint: Comparison with OpenAPI spec
• Details: Request/response for inspection

4. make_repro

Generate cURL command to reproduce API call.

{
  "logIndex": 0,
  "includeHeaders": true,
  "pretty": true
}

Returns:

• cURL Command: Ready to copy-paste
• HTTPie Alternative: Shorter syntax
• JavaScript fetch: For automated tests
• Metadata: Method, URL, headers count

Quick Start

Scenario 1: Debugging a 400 Bad Request

# 1. Load your OpenAPI spec
Use load_openapi with path to your spec file

# 2. Export HAR from browser DevTools
# Network tab → Right-click → Save as HAR with content

# 3. Ingest the logs
Use ingest_logs with HAR file path

# 4. Analyze the failure
Use explain_failure on the failed request

# 5. Get a working cURL command
Use make_repro to generate test command

Example Output:

 ANALYSIS
Status: 400 Bad Request
Severity: HIGH

 ROOT CAUSE
Missing required field: "email"

OpenAPI spec requires:
- name (string, required)
- email (string, format: email, required)

Your request only included "name".

 SUGGESTED FIXES
1. Add "email" field to request body
2. Ensure email format is valid ([email protected])

 TEST COMMAND
…

Scenario 2: Understanding 401 Unauthorized

# 1. Ingest logs showing 401 errors
Use ingest_logs

# 2. Explain the failure
Use explain_failure

# Get specific guidance on:
# - Missing Authorization header
# - Invalid/expired token
# - Wrong authentication scheme
# - Missing scopes/permissions

Slash Command

Use the /debug-api command for a guided debugging workflow:

/debug-api

This activates a systematic 4-step process:

1. Load API documentation (OpenAPI spec)
2. Ingest HTTP logs (HAR or JSON)
3. Analyze failures (explain errors)
4. Generate test commands (cURL repros)

AI Agent

The api-expert agent specializes in API debugging:

• Root cause analysis of HTTP errors
• OpenAPI spec interpretation
• Severity assessment (critical → low)
• Actionable fix suggestions
• Test command generation

Activate by asking questions like:

• "Why is my API returning 400?"
• "Help me debug this authentication error"
• "Analyze these API logs"