Quickstart
Verixa needs Python 3.11+. It has no runtime dependencies, no frontend build step, and needs no API key for the demo.
Run the demo
git clone https://github.com/zyvorai/verixa && cd verixa
VERIXA_TOKEN=Admin@321 python3 -m verixa serve --demo
Open http://127.0.0.1:8788 and sign in as admin / Admin@321. Without VERIXA_TOKEN, a
random token is printed at startup; sign in as admin with that token as the password.
The demo seeds 11 real simulator runs: ten reference-agent runs and one known regression. The reference and regression agents are scripted fixtures, not LLMs.
Use the CLI
python3 -m verixa list # bundled scenarios
python3 -m verixa run all --junit artifacts/reference.xml # whole suite, JUnit for CI
python3 -m verixa run ticket-injection --agent regression # exits 3
python3 -m verixa run refund-commit-timeout --output artifacts/run.json
| Exit code | Meaning |
|---|---|
0 | Every rehearsal passed |
1 | Input or runtime error |
3 | Failed outcome or regression gate |
Record, replay, compare
python3 -m verixa export RUN_ID --output artifacts/evidence.json
python3 -m verixa export RUN_ID --trace --output artifacts/trace.json
python3 -m verixa run refund-happy --agent replay --trace artifacts/trace.json
python3 -m verixa verify artifacts/evidence.json
python3 -m verixa compare BASELINE_ID CANDIDATE_ID
Replay runs the exact action sequence in fresh simulated state; it does not call the original model.
Test a real model
Any endpoint that implements the OpenAI-compatible chat-completions tool-calling contract works. The model proposes calls; Verixa executes them against the simulated world.
export VERIXA_MODEL_URL=http://127.0.0.1:11434/v1
export VERIXA_MODEL=your-tool-capable-model
# export VERIXA_MODEL_KEY=... # when your provider requires it
python3 -m verixa run all --agent model --allow-model-network \
--output artifacts/model.json --junit artifacts/model.xml
Model network calls are opt-in. Task, policy and tool-result text is sent to the configured provider. Next: the console or deploy to a server.