Naja is an open source EDA framework for loading, elaborating, analyzing, optimizing, and transforming hardware designs from RTL SystemVerilog through structural netlists. It is usable from Python or C++.
- SV/Verilog frontend — parse Verilog and elaborate SystemVerilog RTL into a browsable design model
- Netlist analysis — hierarchy, connectivity, equipotentials
- Logic optimization — dead logic elimination, constant propagation
- ECO transformations — direct netlist editing
- Serialization — SNL interchange format (Cap'n Proto) and Verilog output
The best entry point is the najaeda Python package:
pip install najaedaFull documentation: najaeda.readthedocs.io
For AI-assisted design exploration, naja-scope
is a najaeda-based MCP server that gives MCP-compatible assistants a precise,
structured view of elaborated SystemVerilog designs. Instead of pasting large
RTL files into chat, agents can ask targeted questions — what drives a signal,
what is inside a module, where a net comes from — and get small, exact answers
with file-and-line references.
Six hands-on notebooks — open any of them in Colab with no local install needed:
| # | Topic | Colab |
|---|---|---|
| 1 | Getting started — load Verilog, navigate hierarchy, visualize | |
| 2 | Liberty primitives — load a synthesised design with standard cells | |
| 3 | Editing a netlist — rename, disconnect, reconnect, delete | |
| 4 | SystemVerilog elaboration — load and browse an elaborated SV design | |
| 5 | ibex RISC-V core — explore a real-world SV core, collect stats | |
| 6 | Fanout analysis — compute fanout for every net, trace drivers, export to pandas | |
naja_edit is a command-line tool for optimizing and translating netlists.
📺 Presented at ORConf 2024.
# Translate Verilog → SNL
naja_edit -f verilog -t snl -i input.v -o output.snl
# Parse SystemVerilog with explicit top
naja_edit -f systemverilog -t verilog -i input.sv -o output.v --sv_top top
# Dead logic elimination
naja_edit -f snl -t snl -i input.snl -o output.snl -a dle
# Chain optimizations with Python scripts
naja_edit -f snl -t snl -i input.snl -o output.snl -a dle -e pre.py -z post.pyAvailable optimizations (-a): all (DLE + constant propagation + primitives), dle.
Python script examples: src/apps/naja_edit/examples
Regression suite: naja-regress
Ubuntu:
sudo apt-get install g++ libboost-dev python3-dev capnproto libcapnp-dev libtbb-dev pkg-config bison flexmacOS (Homebrew):
brew install cmake capnp tbb bison flex boost
export PATH="/opt/homebrew/opt/flex/bin:/opt/homebrew/opt/bison/bin:$PATH"Nix:
nix-shell -p cmake boost python3 capnproto bison flex pkg-config tbb_2021_8git clone --recurse-submodules https://github.com/najaeda/naja.git
export NAJA_INSTALL=<install-dir>
mkdir build && cd build
cmake ../naja -DCMAKE_BUILD_TYPE=Release -DCMAKE_INSTALL_PREFIX=$NAJA_INSTALL
make && make test && make install
# Add to your environment after install:
export PYTHONPATH=$PYTHONPATH:$NAJA_INSTALL/lib/pythonNaja exposes two complementary APIs:
- SNL (Structured Netlist) — full read/write netlist representation
- DNL (Dissolved Netlist) — fast, read-only flattened view for parallel analysis
Extended documentation: naja.readthedocs.io
C++ snippet: NLUniverseSnippet.cpp
App template (copy to start a new tool): src/app_snippet
- Chat: Matrix #naja:fossi-chat.org
- Bugs / features: GitHub Issues
- Contact: contact@keplertech.io
⭐ If you find Naja useful, starring the repo helps spread the word.
Supported by NLNet through the NGI0 Entrust Fund.