The knowledge layer for reverse engineering

Every binary,understood.

SymGraph is the shared intelligence platform behind GhidrAssist, BinAssist, and IDAssist. It turns reverse engineering work into reusable knowledge: symbols, semantic graphs, analysis workflows, and tool-ready context that compounds instead of disappearing between sessions.

Shared symbol store
Versioned symbols and provenance-aware contributions
Semantic graph layer
Behavior-oriented graph understanding across binaries
MCP-native workflows
External AI clients can drive your analysis surface
See It In Action
SymGraph · Graph

Why teams stall

Stop Context-Switching. Start Understanding.

Reverse engineering is slow because the tools don't think with you. You copy pseudocode into a chat window, wait for a response, then manually apply the results back. You rename one function at a time. You build mental models of call graphs that evaporate between sessions. Dozens of researchers may analyze the same binary with little or no information sharing between them leading to significant lost productivity. Sharing analysis intelligence amongst the RE community can significantly reduce the lead time from discovery to understanding as well as reduce duplication of effort.

The SymGraph service combined with our plugins eliminates that friction. Ask questions about code without leaving your disassembler. Get function explanations, rename suggestions, and vulnerability assessments in seconds. Build a persistent knowledge graph that captures what every function does, how they relate, and where the risks are - so your analysis compounds instead of starting over. Share results with your team, organization and the broader RE community for massive productivity gains.

Plugin Ecosystem

One product surface, native to every major RE tool.

Install in the disassembler you already use. Keep the workflow local. Let SymGraph provide the shared memory, graph intelligence, and platform layer underneath.

GhidrAssist

Ghidra Plugin · Java

A seven-tab AI workbench for Ghidra with explain, query, actions, semantic graph, RAG, and agentic analysis workflows.

  • One-click function explanations with security scoring
  • Context-rich query workflows with code-aware macros
  • Autonomous multi-step investigations through MCP tools

BinAssist

Binary Ninja Plugin · Python

A native Binary Ninja sidebar with streaming LLM output, action suggestions, graph workflows, and controlled extended reasoning.

  • Dockable Qt experience that feels native to Binary Ninja
  • Confidence-scored rename and retype suggestions
  • Deep investigations with adjustable reasoning effort

IDAssist

IDA Pro Plugin · Python

An IDA Pro companion focused on graph exploration, rapid binary triage, and persistent semantic understanding.

  • Visual graph exploration with community detection
  • Structured suggestions for names, types, and comments
  • Persistent local knowledge that compounds between sessions

MCP Servers

Let AI drive your disassembler with the right tool access.

Each plugin ships with a companion MCP server so external LLM clients can navigate functions, read code, annotate findings, and execute structured workflows inside your preferred reverse-engineering environment.

GhidrAssistMCP

Ghidra MCP Server

A Ghidra-native MCP server with rich tool coverage and automatic focus on the active program context.

  • Expose tool-native reverse-engineering actions to external AI clients
  • Support guided, multi-step investigations across live analysis context
  • Pair directly with the corresponding desktop workflow and docs

BinAssistMCP

Binary Ninja MCP Server

A Binary Ninja MCP server designed for multi-binary workflows, prompt-guided investigation, and tool-rich automation.

  • Expose tool-native reverse-engineering actions to external AI clients
  • Support guided, multi-step investigations across live analysis context
  • Pair directly with the corresponding desktop workflow and docs

IDAssistMCP

IDA Pro MCP Server

An IDA Pro MCP server that exposes analysis capabilities safely to external LLM clients and agentic workflows.

  • Expose tool-native reverse-engineering actions to external AI clients
  • Support guided, multi-step investigations across live analysis context
  • Pair directly with the corresponding desktop workflow and docs

Workflow

A reverse-engineering workflow that compounds.

01

Explain Functions

Start at the function level. SymGraph-powered clients turn raw pseudocode into readable summaries, purpose statements, and security context without forcing you into a separate browser tab.

02

Investigate With Agents

When one function is not enough, the MCP stack lets AI clients traverse xrefs, call graphs, data flows, and surrounding context to answer broader reverse-engineering questions.

03

Apply Structured Changes

Rename, retype, annotate, and organize what the model learns. Suggestions stay grounded in the analysis surface so you can review and apply them at analyst speed.

04

Build Shared Intelligence

SymGraph turns one-off analysis into a reusable knowledge layer: symbols, graph structure, and higher-level understanding that can be queried again later or shared across teams.

Build with us

Launch with better defaults for modern reverse engineering.

Browse binaries, read the docs, or start a free account to test the full SymGraph workflow with the desktop clients and API.