This release is a pre-release and may not be stable for production use.
E-commerce Eval
Business acceptance testing for everyday e-commerce agents.
One business acceptance standard, multiple execution methods and verifiable evidence. An agent saying "uploaded successfully" does not prove that the product data is correct or that the approved revision was actually used.
- 32 starter cases across eight directions: intent, company rules, business dependencies, parameters, artifact review, conversation, recovery and safety.
- A local visual platform: inspect traces, business conditions, artifacts, approvals, execution receipts, metrics and missing evidence.
- An Onboarding Skill: investigate a product and draft its cases, contracts and evidence gaps. The Skill is installed separately from GitHub.
Install and run
Requires Python 3.10+. This is the 0.3.0rc2 Alpha candidate, not a stable release. In a virtual environment, install the exact candidate:
python -m pip install "commerce-agent-eval==0.3.0rc2"
commerce-eval demo
Open http://127.0.0.1:8770. The built web UI and demo data are included. Git and Node.js are not required for package installation. The no-key demo does not call a real model, run the entire bank against an agent or write to a production store.
The Python distribution is named commerce-agent-eval; its command is
commerce-eval and its Python module is commerce_eval.
What acceptance means
Business conditions and explicit constraints decide the result. Tool names and counts are diagnostic by default, so packaged business tools and lower-level file tools can be judged against the same business conditions. Required evidence that is absent remains unverifiable, never an automatic pass.
Real integrations still need a controlled execution environment and evidence collectors. No universal zero-adaptation integration or production success rate is claimed. External model calls are explicit opt-in and may transmit inputs to the configured provider and incur charges.
Documentation
1 maintainer
Metadata
1 maintainer
Release files for commerce-agent-eval 0.3.0rc2
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| commerce_agent_eval-0.3.0rc2.tar.gz | 422.4 kB | With provenance | Details for commerce_agent_eval-0.3.0rc2.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| commerce_agent_eval-0.3.0rc2-py3-none-any.whl | Python 3 | none | any | Details for commerce_agent_eval-0.3.0rc2-py3-none-any.whl |
Total release size: 904.3 kB
Release files / commerce_agent_eval-0.3.0rc2.tar.gz
| Download URL | commerce_agent_eval-0.3.0rc2.tar.gz |
|---|---|
| Size | 422.4 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
f3d4836dd9ed3670589c4be7dbf2a1bbe8573b87caea5418125ca4be153eeaf2
|
|
BLAKE2b-256 checksum How to use checksums |
d90e848ff210d56f6bd12c2fb403602de3c29f721f2c69c6a9e584ac90e7fbfd
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Sep 12, 2026.
Transparency logRelease files / commerce_agent_eval-0.3.0rc2-py3-none-any.whl
| Download URL | commerce_agent_eval-0.3.0rc2-py3-none-any.whl |
|---|---|
| Size | 481.9 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
7d72da64006cd5a8de0817f9a3656896ea3ba10728348dda3a420cd82049abe3
|
|
BLAKE2b-256 checksum How to use checksums |
b06694db06e473fd28482d1637da9b7d6c90edc4fca27df603c644188bfca749
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Sep 12, 2026.
Transparency logSecurity for commerce-agent-eval 0.3.0rc2
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
Full provenance
All 2 files in this release have been uploaded with provenance metadata.
Attestation for commerce_agent_eval-0.3.0rc2.tar.gz
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Sep 12, 2026.
Transparency logA note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.
