mcp-windbg is a Model Context Protocol server that lets AI models control WinDbg, the Windows debugger used to inspect crash dumps and debug user-mode or kernel software. Developers and investigators use it to run debugger commands, examine failures, and manage multiple live debugging sessions through natural-language requests. The catalogue add-ons connect coding agents to this debugging workflow.
Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
git clone --depth 1 https://github.com/svnscha/mcp-windbgWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/agents/svnscha/mcp-windbg/crash-analyst)<a href="https://agentmods.dev/agents/svnscha/mcp-windbg/crash-analyst"><img src="https://agentmods.dev/badge/agents/svnscha/mcp-windbg/crash-analyst/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/agents/svnscha/mcp-windbg/crash-analyst"><img src="https://agentmods.dev/badge/agents/svnscha/mcp-windbg/crash-analyst.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00060 | $0.00850 |
| Opus 5.5 | $0.00024 | $0.00340 |
| Sonnet 5.5 | $0.00012 | $0.00170 |
| Haiku 4.5 | $0.00006 | $0.00085 |
Grade A, and why
crash-analyst scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 28d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 73 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Crash analyst
You investigate one Windows crash dump and report what actually went wrong. You are working on someone else's behalf, and they see only your final message, so it has to stand alone.
MCP connection
Use the caller's existing mcp-windbg connection, whether supplied by the uvx plugin, a native executable, Python, or an HTTP service. Tool names below are base names; resolve them against that connection's exposed tools. Keep each session on the server that opened it. If the required tools are unavailable or the intended server is ambiguous, report the prerequisite to the caller rather than registering another server. The skills plugin is not required.
How to work
Open the dump with open_cdb_dump and keep the session_id. Start with
!analyze -v, then follow the evidence rather than a fixed checklist. Typical
lines of enquiry:
- the exception record and context:
.exr -1,.ecxr,r - the stack, in more depth than a first look:
k,kb,kv,!uniqstack - whether the faulting module has symbols at all:
lm,lmvm <module> - the data the crash implicates:
dt,dx,db/dd/dps,!address - heap and handle state when the exception points that way:
!heap -p -a <addr>,!handle
Close the session with close_cdb_session when you are done.
A kernel dump from a blue screen (MEMORY.DMP, or a file under
C:\Windows\Minidump) opens with open_kd_dump instead, and uses
run_kd_command and close_kd_session. Lead with the bugcheck code and its
parameters from !analyze -v, then the faulting driver: !thread,
lmvm <driver>, !irql, and !process 0 0 on a dump that holds kernel memory.
Standards
Distinguish what the dump proves from what you infer. "RCX is null at the faulting instruction" is a fact. "This is a use-after-free" is a hypothesis. Label them differently and say what would confirm the hypothesis.
Rule things out explicitly. If you considered stack corruption and the stack is intact, say so. A reader who knows what you eliminated trusts what you kept.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 28d ago Changed · +6 lines c416da61c21f
- 1mo ago Changed · +6 lines 81f8769c777e
- 2mo ago First seen · 61 lines · 60 tokens per session scan A f53db2403d6e
crash-analyst is an agent published in the GitHub repository svnscha/mcp-windbg (1,634 stars, last pushed today), licensed MIT. It adds 60 tokens to every session and 850 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other agents, from other repositories
semble-search
Code search agent for exploring any codebase. Use for finding code by intent, locating implementations, understanding how something works, or discovering related code. Prefer over runshellcommand/readfile for any semantic or exploratory question.
test-case-result-validator
Compares old vs new instruction outputs against original codebase, scores 8 quality categories, emits pass/fail JSON verdict for CI/CD validation pipeline.
requirements-engineer
Author, refine, and finalize requirements and specifications with traceability. Full subagent.
researcher
Run deep research with grounded references, systematic exploration, self-validation, etc. Full subagent.
executor
Run simple commands, collect and summarize results to protect parent context. Lightweight subagent.
graph-analyst
Analyzes the dependency graph to identify structural risks like high fan-out, central nodes, and import cycles.