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.
Borrowing it
Nothing to install: this file belongs to svnscha/mcp-windbg. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/svnscha/mcp-windbg/develop/.github/prompts/dump-triage.prompt.mdgit 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/commands/svnscha/mcp-windbg/dump-triage)<a href="https://agentmods.dev/commands/svnscha/mcp-windbg/dump-triage"><img src="https://agentmods.dev/badge/commands/svnscha/mcp-windbg/dump-triage/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/commands/svnscha/mcp-windbg/dump-triage"><img src="https://agentmods.dev/badge/commands/svnscha/mcp-windbg/dump-triage.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 | $0.00013 | $0.01414 |
| Opus 5.5 | $0.00005 | $0.00566 |
| Sonnet 5.5 | $0.00003 | $0.00283 |
| Haiku 4.5 | $0.00001 | $0.00141 |
Grade A, and why
dump-triage 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 1mo 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 — 168 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Perform comprehensive analysis of a single Windows crash dump with detailed metadata extraction, call stack analysis, and structured markdown reporting.
WORKFLOW - Execute in this exact sequence:
Step 1: Dump File Identification
If no dump file path provided:
- Ask user to provide the specific crash dump file path, use
list_windbg_dumpsto help them find available dumps.
Step 2: Comprehensive Dump Analysis
Analyze the specified dump file:
Tool: open_windbg_dump
- Parameters:
dump_path: Provided dump file pathinclude_stack_trace: trueinclude_modules: trueinclude_threads: true
The open_windbg_dump tool automatically runs !analyze -v and provides initial analysis output.
If the dump is very large (>5GB) or analysis takes too long we eventually timeout. In that case, inform the user about the timeout and tell him to wait since the analysis keeps running in the background and will complete. The most obvious reason is downloading symbols for running into a timeout here.
Then extract additional metadata with: run_windbg_cmd
- Command 1:
vertarget(OS version and platform details) - Command 2:
lm(loaded modules list) - Command 3:
k(call stack) - Command 4:
.time(dump creation time) - Command 5:
!peb(process environment details) - Command 6:
r(registers)
Then cleanup: close_windbg_dump
Step 3: Generate Structured Analysis Report
REQUIRED OUTPUT FORMAT:
# Crash Dump Analysis Report
**Analysis Date:** [Current Date]
**Dump File:** [filename.dmp]
**File Path:** [Full path to dump file]
## Executive Summary
- **Crash Type:** [Exception type - Access Violation, Heap Corruption, etc.]
- **Severity:** [Critical/High/Medium/Low]
- **Root Cause:** [Brief description of the identified issue]
- **Recommended Action:** [Immediate next steps]
## Dump Metadata
- **File Size:** [MB]
- **Creation Time:** [Date/Time from .time command]
- **OS Build:** [Windows version and build from vertarget]
- **Platform:** [x86/x64/ARM64]
- **Process Name:** [Process name and PID]
- **Process Path:** [Full executable path]
- **Command Line:** [Process command line arguments]
- **Working Directory:** [Process working directory]
## Crash Analysis
**Exception Details:**
- **Exception Code:** [0xC0000005, etc.]
- **Exception Address:** [0x12345678]
- **Faulting Module:** [module.dll or module.exe]
- **Module Version:** [File version]
- **Module Timestamp:** [Build timestamp]
- **Module Base Address:** [0x12345678]
**Call Stack Analysis:**
Frame Module!Function+Offset Parameters
===== ================================================== ==================
[0] module!function+0x123 param1, param2, param3
[1] module!function+0x456 param1, param2
[2] module!function+0x789 param1
[3] module!function+0xabc
[4] module!function+0xdef
[5] [Continue with full stack...]
**Thread Information:**
- **Crashing Thread ID:** [Thread ID]
- **Thread Count:** [Total threads]
- **Other Notable Threads:** [List any threads of interest]
## Technical Details
**Memory Information:**
- **Virtual Size:** [Size from !peb]
- **Working Set:** [Size]
- **Heap Information:** [Heap details if relevant]
**Loaded Modules Summary:**
| Module | Base Address | Size | Path |
|--------|--------------|------|------|
| [Primary modules of interest] | | | |
**Environment Details:**
- **Current Directory:** [From !peb]
- **Environment Variables:** [Key variables if relevant]
- **Command Line:** [Full command line from !peb]
## Root Cause Analysis
[Detailed explanation of what caused the crash, including:]
- **What happened:** [Technical description of the failure]
- **Why it happened:** [Analysis of contributing factors]
- **Code location:** [Specific function/line if identifiable]
- **Memory state:** [Description of memory corruption, null pointers, etc.]
## Recommendations
### Immediate Actions
1. [Specific action item 1]
2. [Specific action item 2]
3. [Specific action item 3]
### Investigation Steps
1. [Follow-up analysis steps]
2. [Code review recommendations]
3. [Testing scenarios to reproduce]
### Prevention Measures
1. [Code changes to prevent recurrence]
2. [Additional validation/checks needed]
3. [Process improvements]
## Priority Assessment
**Severity:** [Critical/High/Medium/Low]
**Justification:** [Explanation of why this severity was assigned based on:]
- Impact on users/system stability
- Frequency of occurrence (if known)
- Ease of reproduction
- Potential for data loss or security issues
## Additional Notes
[Any other relevant observations, similar crashes seen, or context that might be helpful for developers]
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.
- 1mo ago First seen · 168 lines · 13 tokens per session scan A 732dcc48a9db
dump-triage is a command published in the GitHub repository svnscha/mcp-windbg (1,632 stars, last pushed yesterday), licensed MIT. It adds 13 tokens to every session and 1,414 once invoked, about $0.0001 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 2026-09-10.
Other commands, from other repositories
api-aqa-flow-test-correction
Phase 7 Test Corrections of api-aqa-flow (USER APPROVAL REQUIRED).
ui-aqa-flow-test-correction
Phase 8 Test Correction of ui-aqa-flow.
ui-aqa-flow-test-report-analysis
Phase 7 Test Report Analysis of ui-aqa-flow.
api-aqa-flow-execution-and-report-analysis
Phase 6 Execution & Report Analysis of api-aqa-flow (USER INTERACTION REQUIRED).
url
Scrape and integrate documentation from provided URLs using Firecrawl tools.
package
Look up library documentation using Context7 MCP tools for a specific task.