Entries tagged :: mcp-windbg

Build an autonomous crash triage workflow with the Claude Agent SDK, mcp-windbg, and Markdown history.

A standalone, code signed mcp-windbg.exe for machines without Python, and a fix for remote sessions whose target is already running when you connect.

Failed opens no longer leak debugger sessions, a timed-out command on a dump gets cancelled instead of wedging the session, and completion markers are unique per session. Plus what I learned measuring six community pull requests against a real debugger.

A quick follow-up to 1.3.1: open kernel crash dumps with open_kd_dump, and load your debugger extensions automatically with init commands.

Break-in while a command runs, partial output on timeouts, and a debugger that exits is reported right away. Three of the four fixes came from the community.

The MCP server, Claude Code skills, and crash-analysis agent are now separate plugins for use with your preferred installation.

A fix for missing Unicode output on Chinese, Japanese, and Korean Windows systems.

mcp-windbg now ships as a Claude Code plugin with four debugging skills and a crash-analysis agent.

Support for the 2.x MCP SDK, a new wait_for_break tool, and fixes for 'g' freezing the machine it was supposed to release.

Kernel debugging, session IDs, and a redesigned tool surface. The first stable release.

Break into live sessions, redact sensitive output before it reaches the model, and a brand-new documentation site. A year and a month in, with 1.4k stars and a community that keeps showing up.

Adding Prompts API support and streamable-http transport to mcp-windbg.

AI-powered crash analysis gets a major upgrade with PyPI availability, live debugging sessions, and community-driven improvements. One-command installation is finally here.

Because manually squinting at hex dumps is so last century. Let me show you how AI-assisted debugging is leaving WinDBG's command line in the dust.