English Français العربية

Kdb

Community Developer tools

Time-travel debugger with bidirectional execution replay, hash-chain audit logging and SIMD-accelerated stack unwinding.

Source code Documentation

What can you do with Kdb?

  • debugger_attach - Connect to a process
  • debugger_set_breakpoint - Set breakpoints
  • debugger_continue - Resume execution
  • debugger_step_forward - Step forward
  • debugger_step_backward - ⏪ Step backward in time
  • debugger_get_stack_trace - Get stack trace
  • debugger_get_variables - Read memory
  • debugger_export_trace - Export execution trace

Install

The installation steps are in the documentation below. Documentation

Documentation

From the project's README on GitHub, refreshed weekly. Links open on GitHub.

Contents
  1. ✨ KDB - The Kindly Debugger ✨
  2. 🚀 Quick Start (30 seconds)
  3. ⏱️ What It Does
  4. 💜 How It Works
  5. 🏷️ Pricing
  6. 🛠️ MCP Tools
  7. 💬 Example
  8. 📞 Support

✨ KDB - The Kindly Debugger ✨

Give your AI the superpower of traveling back in time to find what went wrong and fix the timeline.

🌐 Website • 🎉 Start Free • 📖 Docs

🚀 Quick Start (30 seconds)

Option 1: One-Line Setup
npx -p @kindly-software-inc/kdb kdb-configure --auto

Follow the prompts to sign in with Google and auto-configure your MCP client.

Option 2: With License Key
npx -p @kindly-software-inc/kdb kdb-configure --auto --license "YOUR-LICENSE-KEY"
Option 3: Manual Config

Add to your MCP client config (Claude Code, Cursor, etc.):

{
  "mcpServers": {
    "kdb": {
      "command": "npx",
      "args": ["@kindly-software-inc/kdb"]
    }
  }
}

That's it! 🎊 Ask your AI: "Debug my crashing program"

⏱️ What It Does

Time-travel debugging for AI workflows. Step forward. Step backward. Debug as if the bug never existed.

FeatureDescription
⏪ Step BackwardRewind execution to find root causes
⏩ Step ForwardResume from any checkpoint
🎯 BreakpointsHardware-level with hit counting
📚 Stack TracesSIMD-accelerated unwinding
🔐 Audit TrailHash-chain integrity for compliance

💜 How It Works

┌─────────────────┐     MCP      ┌─────────────────┐
│  Your Machine   │◄────────────►│  Kindly Server  │
│  (any OS)       │              │  (Linux x86_64) │
│                 │              │                 │
│  Claude Code    │              │  🔮 Time-Travel │
│  Cursor         │              │  🎯 Breakpoints │
│  Any AI         │              │  📊 Analysis    │
└─────────────────┘              └─────────────────┘

No local installation. Works on macOS, Windows, Linux via MCP.

🏷️ Pricing

TierPriceSessions
🆓 HobbyFree5/month (unlimited during 7-day trial!)
💎 Pro$19/mo100/month + unlimited time-travel
🔧 Engineer$49/mo500/month + memory replay + LSH bug search
👥 Teams$129/mo2,000/month + team audit logs
🏢 EnterpriseCustomUnlimited + SOX/SOC2/GDPR/HIPAA compliance

Start Free →

🛠️ MCP Tools

Your AI gets these debugging superpowers:

  • debugger_attach - Connect to a process
  • debugger_set_breakpoint - Set breakpoints
  • debugger_continue - Resume execution
  • debugger_step_forward - Step forward
  • debugger_step_backward - ⏪ Step backward in time
  • debugger_get_stack_trace - Get stack trace
  • debugger_get_variables - Read memory
  • debugger_export_trace - Export execution trace

💬 Example

You: "My Rust program crashes at runtime. PID is 12345"

AI:

I'll attach and debug that for you.

📎 Attached to process 12345
📸 Captured snapshot
🔍 Stack trace shows crash at unwrap() on None in process_data()

The issue is at line 47 in src/main.rs. Here's the fix:
[code suggestion]

📞 Support

Built with 💜 by Kindly Software

Frequently asked questions

What is Kdb?

Time-travel debugger with bidirectional execution replay, hash-chain audit logging and SIMD-accelerated stack unwinding.

How do I install Kdb?

Follow the installation steps in its documentation, shown on this page and in its repository.

Is Kdb open source?

Its code is public on GitHub, but the repository does not declare a standard license.

Who maintains Kdb?

It is published by kindly-software. AI.tn lists it but does not maintain or endorse it.

From the community awesome-mcp-servers lists and the project's GitHub repository. Listing is not an endorsement: review a server's code and permissions before giving it access to your data.