2.0.4

agesafetch

A tool for obtaining your firmware's embedded AGESA version on Linux.

Installation

Binaries

Pre-compiled and signed binaries are provided with all GitLab releases.

From Source

cargo install agesafetch

[!note]

Cargo installs binaries in ~/.cargo/bin by default, so that directory must be in your $PATH for agesafetch to be found. For security, however, not all methods of privilege escalation preserve the $PATH variable, such as sudo configured with a secure_path.

When in doubt, reference the exact path: sudo ~/.cargo/bin/agesafetch

From PyPI

sudo pipx install --global agesafetch
# or, to run it directly without a persistent installation:
sudo pipx run agesafetch

Python Bindings

On top of the agesafetch command, the Python package also provides basic bindings for invoking the AGESA search from your own code:

import agesafetch

version: agesafetch.AGESAVersion | None = agesafetch.find_agesa_version()

[!note]

Like the agesafetch command, this example requires elevated privileges.

Usage

agesafetch [-h]

^{🔒 The AGESA search requires elevated privileges or suitable capabilities.}

Simply run agesafetch to invoke a search for the AGESA version in memory:

$ sudo agesafetch
:: Searching Reserved region #1 (1667 KiB)...
-> Found AGESA version: CezannePI-FP6 1.0.1.1

When run non-interactively, such as in pipes or redirections, agesafetch will automatically suppress all output except for the found version:

$ sudo agesafetch > found_version
$ cat found_version
CezannePI-FP6 1.0.1.1

Testing

agesafetch has been confirmed to work on a broad set of systems, including:

• Various desktop motherboards with
  • B450, B550, and B650 chipsets
  • X570, X670E, and X870E chipsets
• Lenovo ThinkPad P14s Gen 1 & 2 AMD
• An assortment of EPYC-based servers

See Tested Platforms for the complete list. If you tested agesafetch on a new system and want to add it, get in touch!

Author

• Benedikt Vollmerhaus

Thanks To

• Matthias Bräuer, for testing and general advice.
• Tan Siewert, for extensive testing on EPYC-based servers.

License

This project is licensed under the MIT license. See the LICENSE file for more information.

No assets in this project (code, documentation, or anything else) were created by generative AI.