Time-travel debugging over MCP with reverse execution, breakpoints, stack traces, variable inspection, and trace export.
Copy the AI prompt to install this server into Claude Code, Cursor, or another agent β or use 1-click editor setup below.
We haven't yet run this listing's install command through our automated sandbox check. This isn't a red flag β we're steadily working through the catalog.
π‘ Paste the JSON block into your client's configuration file under mcpServers, then restart the application.
Inspect callable tools, capabilities, and parameters exposed to AI agents by Kdb.
kindly-software/kdb MCP server exposes debugging operations to an AI client through the Model Context Protocol. The documented workflow starts by attaching to a process, capturing a debugging snapshot, and examining execution state. From there, an agent can investigate a failure, move through recorded execution, inspect variables, and export a trace.
The server is designed for time-travel debugging rather than only forward stepping. Its listed capabilities include moving backward to investigate root causes, moving forward from a checkpoint, setting breakpoints with hit counting, obtaining stack traces, reading memory through variable inspection, and producing an execution trace for later use. The README also describes hash-chain integrity for its audit trail and SIMD-accelerated stack unwinding.
The MCP client runs on the developerβs machine and communicates with a Kindly server over MCP. The README depicts the Kindly server as running on Linux x86_64 while supporting client machines using macOS, Windows, or Linux. It states that no local installation of the debugger is required.
An AI client can call the debugging tools as it works through a process. A typical investigation may attach to a process, set a breakpoint, continue execution, inspect the stack and variables, then step backward after a failure appears. The example in the README uses a process ID from a crashing Rust program, but the material does not define a language-specific compatibility list.
The documented package is @kindly-software-inc/kdb. The automatic setup command uses npx to run kdb-configure --auto; the setup flow asks the user to sign in with Google and configure an MCP client. A license key can be supplied to that command, but the README also describes a free Hobby tier.
For manual configuration, add an MCP server named kdb with npx as the command and the package as its argument. The README specifically lists Claude Code and Cursor among the clients that can be configured. The package setup and the hosted debugging architecture are described separately, so deployment details beyond the provided MCP configuration are not specified.
The documented MCP tools are:
debugger_attach to connect to a process.debugger_set_breakpoint to create a breakpoint.debugger_continue to resume execution.debugger_step_forward to advance from the current point.debugger_step_backward to rewind execution.debugger_get_stack_trace to retrieve the current stack trace.debugger_get_variables to read variables or memory state.debugger_export_trace to export the execution history.The feature table also mentions hardware-level breakpoints with hit counting, SIMD-accelerated stack unwinding, and hash-chain logging for audit integrity.
kindly-software/kdb MCP server uses a usage-based plan structure. The README lists a free Hobby plan with five sessions per month, while higher tiers add more sessions and features such as unlimited time travel, memory replay, LSH bug search, team audit logs, or compliance support. The material does not state whether every listed debugging feature is available on the free tier.
The documentation identifies the server side as Linux x86_64 and does not provide requirements for the target process, supported programming languages, or debugger permissions. Google sign-in is described for automatic setup, and a license key is supported, but the exact authentication behavior of the MCP calls is not documented. The Discord support link is marked as coming soon in the README.
Factual signals from GitHub, npm, and our automated checks β not a rating.
No reviews yet β be the first to share how this listing worked for you.
Showcase your server listing on GitHub or your project documentation. Embed this dynamic SVG badge to highlight official listing status and live engagement.
[](https://allmcps.com/mcp/kindly-software-kdb)<a href="https://allmcps.com/mcp/kindly-software-kdb"><img src="https://allmcps.com/api/badge/kindly-software-kdb?style=directory" alt="Kdb on AllMCPs" /></a>