Skip to content
Hermes Rubric 1.1.1 · Python 3.10+ · Apache-2.0 published-release licenseOpen source · inspectable boundary

Evidence-first assessment

Assess agent outputs against cited evidence.

Hermes Rubric builds structured rubrics, binds scores to cited evidence, reports coverage facts, and leaves the next action to the caller.

View on GitHub ↗Install from PyPI ↗
Install and runPython
$ pip install hermes-rubric==1.1.1

$ hermes-rubric --intent "Evaluate publication readiness" --target paper.md
Release
v1.1.1
Published on PyPI
Runtime
Python 3.10+
Apache-2.0 published-release license
Quick start
One command
Inspect the repository first
Scope
Bounded tool
Not a system-level guarantee
§ 01

The problem

A bounded operational gap

Why this exists

Make the evidence boundary visible.

A score cannot support an assessment when its criteria, cited evidence, or coverage are unclear. Hermes Rubric assembles those materials into an inspectable result while leaving the next action to the caller.

§ 02

What it does

Read the source and documented interface
01

Structured rubrics

Synthesizes task-specific criteria or accepts frozen and bundled rubrics.

02

Cited scoring

Scores against accepted evidence and preserves evidence, coverage, and receipt data.

03

Caller-controlled feedback

Produces feedback without imposing a release threshold or changing a runtime.

Quick start

Start with the documented command.

Install the current package, then inspect its output and configuration before relying on it in an application or workflow.

Assess a file from the command linelocal command
$ pip install hermes-rubric==1.1.1
$ hermes-rubric --intent "Evaluate publication readiness" --target paper.md
§ 03

Source, package, and license

Primary release links
SOURCEGitHub repositoryInspect the public source and release history.PACKAGEPyPI packagepip install hermes-rubric==1.1.1LICENSEApache-2.0License for the current PyPI release.
§ 04

Related research

Direct Hermes sources
RESEARCHPrecise Records, Unstable MeaningsThe preprint lists Hermes Rubric as evidence-first structured scoring; it is not a validation study of the package.
§ 05

Related concepts

Canonical glossary terms
GLOSSARYLLM-as-a-judge reliabilityHow consistently model-based judges produce defensible assessments under defined conditions.GLOSSARYInspection coverageWhether the inspection layers required for a decision actually ran.GLOSSARYEpistemic EngineeringEngineering how AI systems handle evidence, uncertainty, and justification.GLOSSARYTelemetry-to-Claim GateA proposed framework for recording the evidentiary basis of a telemetry-based claim.

The boundary

Useful evidence is not a blanket guarantee.

The aggregate is a signal, not a verdict. Hermes Rubric does not impose a binary release gate or decide what an application does next.

Compare it with the other open-source tools →