---
type: "post"
id: "ghidramcp-7-0-0-stable-release-ships-with-56633c"
slug: "ghidramcp-7-0-0-stable-release-ships-with-56633c"
title: "GhidraMCP 7.0.0 stable release ships with 209 MCP tools for Ghidra 12.1.4"
description: "GhidraMCP 7.0.0 is now a stable release of an MCP server for the Ghidra reverse-engineering platform."
retrieval_nugget: "GhidraMCP 7.0.0 is now a stable release of an MCP server for the Ghidra reverse-engineering platform. It packages 209 tools behind a Python bridge and a Ghidra extension, giving teams working on binaries a defined agent-facing surface rather than an ad hoc local integration."
published_at: "2026-10-28"
updated_at: "2026-10-28"
record_date: "2026-10-28"
date_kind: "published_at"
topics: ["mcp"]
entities: ["GhidraMCP","Ghidra","Model Context Protocol","Java"]
related_patterns: ["agent-ready-software-exposes-capabilities","agent-protocols-become-interoperability-layer"]
novelty_label: "new_fact"
what_changed: "GhidraMCP has now reached stable version 7.0.0, with 209 MCP tools for Ghidra 12.1.4."
prior_refs: ["memory:220dec3b-9fdd-8986-a149-84c9dad1b159","memory:b6616268-9b8c-87cc-a195-14339718dfd4"]
source_url: "https://github.com/bethington/ghidra-mcp/releases"
source_urls: ["https://github.com/bethington/ghidra-mcp/releases"]
schema_version: "newruntime-agent-readable-v0.2"
stable_id: "post:ghidramcp-7-0-0-stable-release-ships-with-56633c"
status: "published"
visuals: []
editor: {"role":"editor","name":"Andrey Reshetnikov","profile":"https://newruntime.com/owner-profile.md","profile_json":"https://newruntime.com/owner-profile.json","same_as":["https://www.linkedin.com/in/reshetnikov1/","https://t.me/lapetuse"]}
routes: {"html":"https://newruntime.com/posts/ghidramcp-7-0-0-stable-release-ships-with-56633c/","markdown":"https://newruntime.com/posts/ghidramcp-7-0-0-stable-release-ships-with-56633c.md","json":"https://newruntime.com/posts/ghidramcp-7-0-0-stable-release-ships-with-56633c.json"}
---

# GhidraMCP 7.0.0 stable release ships with 209 MCP tools for Ghidra 12.1.4

## Retrieval answer

GhidraMCP 7.0.0 is now a stable release of an MCP server for the Ghidra reverse-engineering platform. It packages 209 tools behind a Python bridge and a Ghidra extension, giving teams working on binaries a defined agent-facing surface rather than an ad hoc local integration.

## Summary

GhidraMCP 7.0.0 is now a stable release of an MCP server for the Ghidra reverse-engineering platform. It packages 209 tools behind a Python bridge and a Ghidra extension, giving teams working on binaries a defined agent-facing surface rather than an ad hoc local integration.

## What happened

The `bethington/ghidra-mcp` project released GhidraMCP 7.0.0 as its latest stable version on October 21, 2026. The release describes itself as a production-ready Model Context Protocol server for Ghidra, the reverse-engineering platform. Its core proposition is to expose Ghidra functionality through MCP so that an MCP client can invoke a bounded set of reverse-engineering operations through the server rather than treating the desktop application solely as a human-operated interface.

The release reports 209 MCP tools. It separately lists 205 GUI endpoints and 190 headless endpoints, which indicates that the project presents related capabilities through both an installed Ghidra GUI workflow and a non-GUI execution surface. The stated compatibility baseline is Ghidra 12.1.4 and Java 21 LTS. Those version targets matter operationally: the package is not presented as a generic bridge for any Ghidra or Java installation, but as a release tested and packaged around named dependencies.

Installation is distributed in two pieces. The `GhidraMCP-7.0.0.zip` asset is installed in Ghidra through the extension installer; after a restart, the user enables the GhidraMCP plugin and starts the MCP server from Ghidra's Tools menu. The second asset is a Python wheel, `ghidra_mcp_bridge-7.0.0-py3-none-any.whl`. The release instructs users to install that wheel with `uv tool install` or `pip install`, then run `bridge-mcp-ghidra` or its Python module equivalent.

That separation is consequential for adoption. Ghidra hosts the extension and server, while the installed Python bridge provides the MCP protocol-facing command. A team evaluating the release therefore needs to plan for both the local Ghidra setup and the environment in which its MCP client will launch the bridge. The release links to the repository changelog and README for detailed release notes and fuller documentation, but the release page itself does not enumerate every tool or provide a task-level performance comparison.

## What the archive adds

This is a new stable-release fact on a known story. The earlier Field Note, [ghidra-mcp: Ghidra MCP Server](https://newruntime.com/posts/ghidra-mcp-ghidra-mcp-server-analysis-acdbd3/), established the underlying thesis: Ghidra capabilities were being packaged behind an MCP server for agent workflows. Version 7.0.0 adds a declared stable package, concrete endpoint counts, and a named dependency baseline.

## Trend signal

The release supports the pattern that [agent-ready software exposes capabilities](https://newruntime.com/patterns/agent-ready-software-exposes-capabilities/). Instead of asking an agent to operate an unrestricted desktop environment, GhidraMCP advertises a catalog of 209 MCP tools and distinct GUI and headless endpoint counts. It also modestly supports [agent protocols become the interoperability layer](https://newruntime.com/patterns/agent-protocols-become-interoperability-layer/): the bridge is positioned as the connection between an MCP client and a specialized reverse-engineering application.

## Why now

The stable version follows the project's 7.0.0 release-candidate cycle, including a release candidate published on October 14. The final release was published on October 21 and carries a verified GitHub signature for its release commit. For prospective users, that changes the status from a testing release to the repository's latest release rather than introducing a newly described category of capability.

## Why it matters

For teams that analyze binaries, agent integration often fails at the boundary between a language-model workflow and the specialized program that holds the relevant program state. This release provides a packaged boundary: install the extension in Ghidra, run its server, and launch the Python MCP bridge. The explicit tool count is not proof that every operation will be reliable for a particular target, but it does give evaluators a concrete surface to inspect and test.

The headless endpoint count is especially relevant to teams considering repeatable analysis workflows rather than purely interactive use. They can assess whether their desired Ghidra tasks are available through the appropriate surface before designing an agent workflow around them. The Java and Ghidra version requirements should be treated as deployment constraints, not incidental release metadata.

## Limits and uncertainty

The release page establishes that 7.0.0 is available and states its counts, assets, installation sequence, and version targets. It does not list all 209 tools on the page, define which tasks differ between GUI and headless operation, or show benchmark results, reliability rates, security controls, or permission boundaries. “Production-ready” is the project's description, not independent validation. The material also does not identify supported MCP clients, licensing terms, operating-system coverage, pricing, or the practical quality of agent-produced reverse-engineering conclusions.

## Skill to build now

Create a small compatibility check before connecting an agent to a real analysis workflow. Confirm that the target machine runs Ghidra 12.1.4 and Java 21, install the extension and Python bridge in an isolated environment, then verify that the bridge can start and that the needed task appears in the available tool surface. Record whether that task requires the GUI endpoint or can run headlessly. This turns the release's advertised counts into a deployment decision based on the operations your team actually needs.

## Reader takeaway

GhidraMCP 7.0.0 is a stable, packaged MCP integration for Ghidra with 209 advertised tools and separate GUI and headless surfaces. It is a concrete option for binary-analysis teams, but its value depends on whether the specific Ghidra operations they need are exposed, testable, and compatible with their local environment.

## Open question

Which of the 209 advertised tools provide dependable headless workflows for repeatable analysis, and what permissions or safeguards constrain their use in an agent-driven environment?

Editor: [Andrey Reshetnikov](https://newruntime.com/owner-profile.md), AI product and AI engineering; profile for recruiters and agents: https://newruntime.com/owner-profile.json

## Continue in the graph

Next records for an agent, linked to their Markdown twins. Replace `.md` with `.json` for structured fields, including `next_reads`.

- topic hub: [Mcp](https://newruntime.com/topics/mcp.md): Explore the mcp topic hub.
- collection: [Field Notes](https://newruntime.com/posts.json): Open the latest editorial analysis.
- collection: [Source Ledger](https://newruntime.com/source-ledger.json): Inspect the public source evidence graph.
- whole graph: [discovery graph](https://newruntime.com/discovery-graph.json) (typed nodes and edges) and [site index](https://newruntime.com/site-index.md)