angr
angr.io
- Free plan
- Windows
- macOS
- Linux
- Self-hosted
What angr is
angr is a free, open-source Python platform for binary analysis that combines static analysis with dynamic symbolic analysis. Its symbolic execution engine explores program paths and solves constraints on symbolic inputs. angr can recover control-flow graphs, disassemble machine code, lift code to an intermediate language, and use its decompiler to produce angr Intermediate Language or C pseudocode. The platform supports custom analyses, architectures, platforms, exploration techniques, and hooks. Related project libraries cover architecture information, intermediate representations, binary loading, and constraint solving. angr runs on Windows, macOS, and Linux and supports architectures including AArch64, AMD64, ARM variants, AVR8, MIPS, PowerPC, RISC-V 64, S390X, and x86. It is a Python 3.10+ library available through PyPI and installable with pip; source installation and a Docker image are also documented. The project releases it under the permissive BSD license. Documentation presents reversing, vulnerability discovery, firmware analysis, and exploitation as example uses. The GUI, angr-management, is described as being in a very alpha state. Support is provided through Discord, GitHub issues and pull requests, and a public mailing list, with potentially long email response times.
Who it suits
angr is intended for developers and security practitioners working on reversing, vulnerability discovery, firmware analysis, or exploitation. It suits Python users who need programmable symbolic and binary-analysis capabilities across many architectures.
Good to know, and what to watch
- Free BSD-licensed open-source software
- Combines static and dynamic symbolic analysis
- Includes decompiler and control-flow recovery
- Supports many architectures and operating systems
- Extensible analyses, hooks, and exploration techniques
- GUI is described as very alpha
- Requires Python 3.10 or newer
- Open-source email support may take a long time
- No security certifications or compliance attestations stated
Plans
Every plan and priceFree and Open Source
Free plan
The verdict
angr provides a programmable toolkit for symbolic execution, disassembly, lifting, decompilation, and control-flow analysis across broad architecture support. Its command-line and library focus suits technical users; the project’s GUI remains at a very early stage.
What the maker publishes
- Control-flow analysis
- angr provides techniques for recovering control-flow graphs. source
- Decompilation
- Its decompiler can produce angr Intermediate Language and C pseudocode from machine code. source
- Deployment
- The project documents installation from source and provides an angr Docker Hub image. source
- Disassembly and lifting
- angr can disassemble code and lift it to an intermediate language. source
- Extensibility
- The platform supports custom analyses, architectures, platforms, exploration techniques, and hooks. source
- GUI
- The project points users to angr-management as its GUI and describes it as being in a very alpha state. source
- Installation
- angr is a Python 3.10+ library available through PyPI and can be installed with pip. source
- Intended use
- The documentation presents reversing, vulnerability discovery, and exploitation as angr example use cases. source
- Libraries
- The angr project maintains archinfo, pyvex, pypcode, CLE, and Claripy libraries for architecture information, intermediate representations, binary loading, and constraint solving. source
- License
- The project says angr is free and open-source software under the permissive BSD license. source
- Platforms
- The angr site states that it runs on Windows, macOS, and Linux. source
- Product
- angr is an open-source binary analysis platform for Python that combines static and dynamic symbolic analysis. source
- Security and compliance
- The official pages opened describe a BSD-licensed analysis tool but do not state security certifications or compliance attestations. source
- Support
- The project directs users to Discord, GitHub issues and pull requests, or a public mailing list, and describes support as a typical open-source model with potentially long email response times. source
- Symbolic execution
- Its symbolic execution engine explores program paths and solves constraints on symbolic inputs. source
Checked against the maker's pages on 3 Oct 2026.