You cannot select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
 
 
 
 
 
Go to file
patryk4815 b87c8d4b04
Revert "Improving `vmmap` output when debugging kernel (#3020)"
6 months ago
.devcontainer Remove vscode personalization from devcontainer (#2602) 1 year ago
.github Show demo gif on docs front page (#3055) 6 months ago
docs Adding a few helper commands for x86 paging (#3051) 6 months ago
nix Fix terminfo in portable-lldb and fix terminal size (#3050) 6 months ago
profiling More robust uv usage (#2986) 6 months ago
pwndbg Revert "Improving `vmmap` output when debugging kernel (#3020)" 6 months ago
scripts Make it clear to contributors LLDB is needed for doc generation (#3057) 6 months ago
tests Revert "Improving `vmmap` output when debugging kernel (#3020)" 6 months ago
.dockerignore Bringt back Archlinux CI. Fixes #1772, closes #1783 (#1800) 2 years ago
.envrc Add nix development shell (#2166) 1 year ago
.gitignore Cleanup testing scripts a bit (#3041) 6 months ago
CREDITS.md docs: Add note about giving credit to ported tools (#2154) 2 years ago
Dockerfile More robust uv usage (#2986) 6 months ago
Dockerfile.arch More robust uv usage (#2986) 6 months ago
Dockerfile.lldb More robust uv usage (#2986) 6 months ago
LICENSE.md Update LICENSE.md, fix license year (#2660) 11 months ago
README.md A few more docs fixups (#3064) 6 months ago
binja_script.py Configure IDA/Binary Ninja integration server endpoint through environment (#2746) 10 months ago
crossarch-tests.sh Cleanup testing scripts a bit (#3041) 6 months ago
docker-compose.yml Simplify local testing inside Docker environment (#2777) 9 months ago
flake.lock Bump flake.lock + cleanup nfpm.yml (#2817) 8 months ago
flake.nix use nixfmt-tree instead of calling the nixfmt-rfc-style directly (#2928) 8 months ago
gdbinit.py More robust uv usage (#2986) 6 months ago
ida_script.py IDA: Support lookup of stack variables using `$ida("name")` (#2926) 8 months ago
kernel-tests.sh Cleanup testing scripts a bit (#3041) 6 months ago
lint.sh More robust uv usage (#2986) 6 months ago
lldbinit.py More robust uv usage (#2986) 6 months ago
mkdocs.yml Show demo gif on docs front page (#3055) 6 months ago
pwndbg-lldb.py Add programatic controls to the LLDB Pwndbg CLI (#2785) 9 months ago
pyproject.toml Bump version to 2025.05.30 (#3052) 6 months ago
setup-dev.sh Make it clear to contributors LLDB is needed for doc generation (#3057) 6 months ago
setup.sh More robust uv usage (#2986) 6 months ago
tests.sh Cleanup testing scripts a bit (#3041) 6 months ago
unit-tests.sh More robust uv usage (#2986) 6 months ago
uv.lock Bump version to 2025.05.30 (#3052) 6 months ago

README.md

repository-open-graph

pwndbg

license Tests codecov.io Discord

pwndbg (/paʊnˈdiˌbʌɡ/) is a GDB and LLDB plug-in that makes debugging suck less, with a focus on features needed by low-level software developers, hardware hackers, reverse-engineers and exploit developers.

It has a boatload of features, see our Features page and CHEATSHEET (feel free to print it!). If you have any questions you may read the documentation or asks us in our Discord server.

Why?

Vanilla GDB and LLDB are terrible to use for reverse engineering and exploit development. Typing x/30gx $rsp or navigating cumbersome LLDB commands is not fun and often provides minimal information. The year is 2025, and core debuggers still lack many user-friendly features such as a robust hexdump command. WinDbg users are completely lost when they occasionally need to bump into GDB or LLDB.

Pwndbg is a Python module which can be loaded into GDB or run as a REPL interface for LLDB. It provides a suite of utilities and enhancements that fill the gaps left by these debuggers, smoothing out rough edges and making them more user-friendly.

Installation

See installation instructions.

What about ...?

Many past (gdbinit, PEDA) and present projects (GEF, bata24/GEF) offer great features, but are hard to extend and are packaged as large single files (103KB, 195KB, 423KB, 4.12MB). Pwndbg aims to replace them with a faster, cleaner, and more robust implementation.

When to Use GDB or LLDB?

Pwndbg supports both GDB and LLDB, and each debugger has its own strengths. Here's a quick guide to help you decide which one to use:

Use Case Supported Debugger
Debugging Linux binaries or ELF files GDB, LLDB
Debugging Mach-O binaries on macOS LLDB
Linux kernel debugging (qemu-system) GDB, LLDB
Linux user-space emulation (qemu-user) GDB
Embedded debugging (ARM Cortex M* or RISC-V/32) GDB, LLDB

Pwndbg ensures a consistent experience across both, so switching between them is seamless.

The LLDB implementation in pwndbg is still in early-stage and may contain bugs or limitations.
Known issues are tracked in GitHub Issues.

If you encounter any problems, feel free to report them or discuss on our Discord server.

Compatibility Table

Feature Supported Version Notes
pwndbg-gdb - Python 3.10+
- GDB 12.1+
Battle-tested on Ubuntu 22.04/24.04
pwndbg-lldb - Python 3.12+
- LLDB 19+
Experimental/early-stage support
qemu-user QEMU 8.1+ vFile API is needed for vmmap
qemu-system QEMU 6.2+ Supported version since ubuntu 22.04

Contributing

Pull requests are welcome ❤️. Check out the Contributing Guide.

Acknowledgements

Pwndbg was originally created by Zach Riggle, who is no longer with us. We want to thank Zach for all of his contributions to pwndbg and the wider security community.