This release has been yanked by its maintainers, and will be ignored by installers, except when explicitly specified.
Consider using release 1.3.0 instead.
Reason given by maintainers: Old version with a security issue (GHSA-5xgq-4r93-29m3). Use the newest judgekeeper.
judgekeeper ™
Check your LLM-as-a-judge in 5 steps.
Find. Mark. Measure. Fix. Re-check.
An LLM-as-a-judge is an AI model that grades another model's answers. judgekeeper shows how often yours agrees with you, and where it goes wrong.
Install
In your project folder, in its virtual environment:
pip install judgekeeper
judgekeeper
Or, with no virtual environment: uvx judgekeeper. Python 3.11 to 3.14.
Reads promptfoo, DeepEval, Inspect AI, MLflow, or your own CSV, TSV or JSONL file. For Inspect AI logs or MLflow: pip install "judgekeeper[inspect]" or pip install "judgekeeper[mlflow]".
Using a coding agent? Give it this page. It runs the steps and gives you the marking page link; you mark, then it explains the result.
The 5 steps
- Find:
judgekeeperfinds your eval results. - Mark: a page opens in your browser. Mark answers Pass or Fail; the LLM-as-a-judge's grade stays hidden.
- Measure: run
judgekeeperagain to see how often it agrees with you. You need at least 15 Pass and 15 Fail marks. - Fix:
judgekeeper fixhelps you write a better rule and tests it on marks kept aside. - Re-check:
judgekeeper checkshows if it still agrees after you change the model or the rule.
judgekeeper only reads files. It makes no AI call and needs no API key. See SECURITY.md.
The numbers
- TPR: of the answers you passed, how many it passed too.
- TNR: of the answers you failed, how many it failed too.
- Kappa: how much you agree beyond what matching at random would give.
Each comes with a range, such as "about 90% (between 84% and 93%)". The result says "good enough to use" only when the low end of both TPR and TNR is at least 80%. The numbers describe the answers you marked, not a promise about the next ones.
How the numbers are worked out: reference.
Checked on public data
- JailbreakBench: the ranges held the true value in 93 of 100 tests (all three numbers at once).
- RAGTruth: three LLM-as-a-judge setups on news summaries: judgekeeper said "needs work" in all 12 runs. Changing the model did not close the gap.
- RAGTruth question answers and data-to-text: "needs work" in all 4 runs.
- JailbreakBench graded by two other setups: "needs work" for both; each passed more than half of the expert-failed answers it graded (setup B could not grade 61 of 300).
- Empty answers: one model failed 0 of 20 empty summaries that should fail.
- Simulation: the ranges held the true value 96.2 times in 100 on average.
More
A guide on the website walks through the 5 steps. judgekeeper --help lists every command, and the reference explains each one. Made by one person; issues welcome, pull requests are not taken.
License
Apache-2.0 for the code. The name and logo are not covered: see TRADEMARKS.md.
judgekeeper and the judgekeeper logo are trademarks of Sathvik Thota. The Apache-2.0 license covers the code, not the name or logo.
Metadata
Release files for judgekeeper 1.0.1
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 |
|---|---|---|---|---|
| judgekeeper-1.0.1.tar.gz | 450.2 kB | With provenance | Details for judgekeeper-1.0.1.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| judgekeeper-1.0.1-py3-none-any.whl | Python 3 | none | any | Details for judgekeeper-1.0.1-py3-none-any.whl |
Total release size: 930.0 kB
Release files / judgekeeper-1.0.1.tar.gz
| Download URL | judgekeeper-1.0.1.tar.gz |
|---|---|
| Size | 450.2 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
00048f58eb0037edbb7a232b9d8fbe9ca5b370d26ed228fc11709c395843b76a
|
|
BLAKE2b-256 checksum How to use checksums |
f97dd0f4714219fbc9c8e27a4010f09736f3536291ce1c20ea15d2e1482b5972
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.16
|
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 Oct 31, 2026.
Transparency logRelease files / judgekeeper-1.0.1-py3-none-any.whl
| Download URL | judgekeeper-1.0.1-py3-none-any.whl |
|---|---|
| Size | 479.8 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
9aa0e977734206b027b3d4fef932a92af2ae811a161f1f5009c29cca98f0fd7b
|
|
BLAKE2b-256 checksum How to use checksums |
b1f4081ddaaf9328f48d673821fab866799dd9e6e9a23a3dd024e23bf1943143
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.16
|
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 Oct 31, 2026.
Transparency logSecurity for judgekeeper 1.0.1
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
Both files in this release have been uploaded with provenance metadata.
Attestation for judgekeeper-1.0.1.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 Oct 31, 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.