MCP server for .NET assembly analysis and reverse engineering using dnSpy decompiler
Mixed-quality tool definitions. 15 of 19 tools have descriptions and visible parameter schemas with type information. Naming is generally verb-driven (get_, list_, set_, clear_, decompile_, inspect_, find_, search_, resolve_, detect_) and appropriate. However, schemas lack formal constraints (enums, ranges, min/max), parameter descriptions are minimal (1-2 sentences), output schemas are not documented in the visible code, and error handling is inferred but not explicitly shown. The assembly analysis tools are well-scoped and read-only; breakpoint and decompile tools are reasonable. Notable gaps: no pagination support for result-heavy tools (decompile_assembly, list_types, find_methods), no documented output structures for LLMs to plan downstream calls, and no explicit error recovery guidance. The code shows try-catch blocks returning generic error tuples but does not show structured error classification or recovery hints. Overall, definitions are adequate for discovery but lack the polish needed for confident LLM integration at scale.
Clear breakpoints: by ID, by assembly, or all at once
Decompile the entire assembly to C# source code
Decompile a specific method to C# source code
Decompile a specific type to C# source code
Fast static detection of anti-debug techniques via raw binary byte search and PE header inspection. Checks P/Invoke declarations, managed API usage, TLS callbacks, thread hiding, timing checks, hardware breakpoint detection, and exception-based patterns. No decompilation — completes in milliseconds.
Dump IL (CIL) disassembly for the whole assembly, a type, or a specific method
No output schemas documented in tool definitions. LLMs cannot plan multi-step sequences or extract required fields for downstream calls. E.g., get_pe_info, decompile_assembly, inspect_method have complex return types (PEInfo, List<ResourceInfo>, etc.) but no structured schema shown to the agent.
List and search tools (decompile_assembly, list_types, find_methods, search_strings, search_members) lack pagination parameters (offset/limit/cursor) and result caps. Decompiling a large assembly or listing all types in .NET Framework can return thousands of items, exceeding context windows. No documented page size limits.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-22 | D | 52 | 2026-07-28+ | v2 |
| 2026-03-09 | F | 47 | - | v1 |
Find methods in the assembly, optionally filtered by name pattern
Get PE/COFF header information, assembly metadata, and target framework of a .NET assembly
List all manifest resources embedded in the assembly
Inspect a breakpoint in full detail: decompiled source, IL context, stack state, and callers
Inspect a specific method: signature, parameters, decompiled source, and optionally IL code
Inspect a type's structure: fields, methods, properties, interfaces, and optionally its decompiled source
List all virtual breakpoints, optionally filtered by assembly
List all P/Invoke (DllImport) declarations in the assembly
List all type definitions in the assembly
Resolve a metadata token (hex, e.g. '0x02000001') to its definition
Search for types, methods, fields, and properties by name pattern
Search for string literals in the assembly's decompiled source
Set a virtual breakpoint at an IL offset in a method. Returns deep decoded context about that point.
Parameter descriptions lack formal constraints. E.g., ilOffset is described as 'IL byte offset (optional, defaults to 0)' but no range documented (0-N bytes varies by method). pattern and useRegex in search_strings lack enum constraints. No min/max for numeric parameters.
Error handling is present (try-catch blocks visible in code) but not documented in tool descriptions. No guidance on what errors are retryable, what indicates user error, or what IDs to provide in recovery. E.g., if assemblyPath is invalid, what should the LLM do next?
set_breakpoint and clear_breakpoints modify state but descriptions do not explicitly state this. Agents need to know which tools have irreversible side effects for planning and retry logic.