Metadata-Version: 2.5
Name: claimbound-evidence
Version: 0.4.8
Summary: Evidence cards for narrow public AI, ML and data claims: frozen protocol, hashed sources, honest result status.
Project-URL: Homepage, https://github.com/ClaimBound/claimbound-evidence
Project-URL: Repository, https://github.com/ClaimBound/claimbound-evidence
Project-URL: Issues, https://github.com/ClaimBound/claimbound-evidence/issues
Project-URL: Changelog, https://github.com/ClaimBound/claimbound-evidence/blob/main/CHANGELOG.md
Project-URL: Documentation, https://github.com/ClaimBound/claimbound-evidence/tree/main/docs
Project-URL: PyPI, https://pypi.org/project/claimbound-evidence/
Project-URL: Discussions, https://github.com/ClaimBound/claimbound-evidence/discussions
Project-URL: Citation, https://github.com/ClaimBound/claimbound-evidence/blob/main/CITATION.cff
Author: Rostyslav Shcherbyna
License-Expression: Apache-2.0
License-File: LICENSE
License-File: NOTICE
Keywords: ai-transparency,claim-provenance,claim-verification,evidence-cards,open-data,open-science,pre-registration,provenance,reproducibility,source-audit
Classifier: Development Status :: 4 - Beta
Classifier: Environment :: Console
Classifier: Intended Audience :: Developers
Classifier: Intended Audience :: Information Technology
Classifier: Intended Audience :: Science/Research
Classifier: Operating System :: OS Independent
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3 :: Only
Classifier: Programming Language :: Python :: 3.12
Classifier: Programming Language :: Python :: 3.13
Classifier: Programming Language :: Python :: 3.14
Classifier: Topic :: Scientific/Engineering
Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
Classifier: Topic :: Scientific/Engineering :: Information Analysis
Classifier: Topic :: Software Development :: Quality Assurance
Requires-Python: >=3.12
Requires-Dist: numpy>=2.5.2
Requires-Dist: pdfplumber>=0.11
Requires-Dist: pypdf>=6.16.0
Provides-Extra: dev
Requires-Dist: pytest-cov>=6.0; extra == 'dev'
Requires-Dist: pytest-xdist>=3.5; extra == 'dev'
Requires-Dist: pytest>=9.1.1; extra == 'dev'
Requires-Dist: ruff>=0.15; extra == 'dev'
Provides-Extra: docs
Requires-Dist: mkdocs-material>=9.5; extra == 'docs'
Requires-Dist: mkdocs<2,>=1.6; extra == 'docs'
Description-Content-Type: text/markdown

<h1 align="center">
  <picture>
    <source media="(prefers-color-scheme: dark)" srcset="https://raw.githubusercontent.com/ClaimBound/claimbound-evidence/main/docs/assets/logo/claimbound-logo-dark.svg">
    <img src="https://raw.githubusercontent.com/ClaimBound/claimbound-evidence/main/docs/assets/logo/claimbound-logo-light.svg" alt="ClaimBound Evidence" width="420">
  </picture>
</h1>

<p align="center">
  <b>Where is the evidence?</b><br>
  Evidence cards for narrow public AI, ML and data claims: frozen protocol,
  hashed sources, honest result status.
</p>

<p align="center">
  <a href="https://pypi.org/project/claimbound-evidence/"><img alt="PyPI" src="https://img.shields.io/pypi/v/claimbound-evidence?logo=pypi&logoColor=white"></a>
  <a href="https://pypi.org/project/claimbound-evidence/"><img alt="Python >=3.12" src="https://img.shields.io/badge/python-%3E%3D3.12-blue?logo=python&logoColor=white"></a>
  <a href="https://pypistats.org/packages/claimbound-evidence"><img alt="PyPI downloads per month" src="https://img.shields.io/pypi/dm/claimbound-evidence?label=downloads%2Fmonth"></a>
  <a href="https://pepy.tech/project/claimbound-evidence"><img alt="Total PyPI downloads" src="https://static.pepy.tech/badge/claimbound-evidence"></a>
  <a href="https://github.com/ClaimBound/claimbound-evidence/actions/workflows/tests.yml"><img alt="tests" src="https://img.shields.io/github/actions/workflow/status/ClaimBound/claimbound-evidence/tests.yml?branch=main&label=tests&logo=github"></a>
  <a href="https://github.com/ClaimBound/claimbound-evidence/blob/main/LICENSE"><img alt="License: Apache-2.0" src="https://img.shields.io/badge/license-Apache--2.0-blue"></a>
  <a href="https://github.com/ClaimBound/claimbound-evidence/releases/latest"><img alt="Latest release" src="https://img.shields.io/github/v/release/ClaimBound/claimbound-evidence?label=release"></a>
  <a href="https://github.com/ClaimBound/claimbound-evidence/blob/main/docs/registry/evidence_index.json"><img alt="Evidence cards" src="https://img.shields.io/badge/dynamic/json?url=https%3A%2F%2Fraw.githubusercontent.com%2FClaimBound%2Fclaimbound-evidence%2Fmain%2Fdocs%2Fregistry%2Fevidence_index.json&query=%24.card_count&label=evidence%20cards&color=0A97BB"></a>
</p>

<p align="center">
  <img src="https://raw.githubusercontent.com/ClaimBound/claimbound-evidence/main/docs/assets/demo/quickstart.gif" alt="Terminal recording: claimbound validate-card and inspect card on a real evidence card" width="760">
</p>

```bash
pipx install claimbound-evidence        # or: uv tool install claimbound-evidence
claimbound validate-card path/to/card.json
```

ClaimBound turns a narrow public AI, ML, data or software-development claim into
a small evidence card: protocol, source boundary, hashes, exact result status,
claim boundary and reproduction level.

It is not a model leaderboard, hosted scoring service or certification authority.
It is an open-source toolkit for asking one plain question:

```text
Where is the evidence?
```

If there is no evidence card, the statement is still only a claim. Green means
one narrow claim passed under a frozen protocol — not "trust everything".
`PASSED_UNDER_PROTOCOL` is not the same as independently reproduced: most public
cards are still `SINGLE_OPERATOR` / `not independently reproduced` until a
separate rerun records otherwise. Negative, blocked and drift outcomes are
first-class evidence too.

Read the one-screen walkthrough in [ClaimBound in 30 seconds](https://github.com/ClaimBound/claimbound-evidence/blob/main/docs/CLAIMBOUND_IN_30_SECONDS.md),
the [reviewer path](https://github.com/ClaimBound/claimbound-evidence/blob/main/docs/REVIEWER_PATH.md),
the [non-developer start guide](https://github.com/ClaimBound/claimbound-evidence/blob/main/docs/START_WITHOUT_CODING.md) or the
[documentation index](https://github.com/ClaimBound/claimbound-evidence/blob/main/docs/README.md) for reviewer, operator and advanced paths.

![ClaimBound workflow](https://raw.githubusercontent.com/ClaimBound/claimbound-evidence/main/docs/assets/claimbound_workflow.svg)

## What It Is And Is Not

| ClaimBound is | ClaimBound is not |
| --- | --- |
| A card per **one narrow claim**: protocol, source boundary, hashes, result status, claim boundary, reproduction level | A model leaderboard, benchmark ranking or quality score |
| A validator (`validate-card`, `validate-all`) that anyone can rerun locally | A hosted scoring service or certification authority |
| Honest about negative, blocked and drifted results (they are first-class evidence) | A way to declare a system safe, compliant or correct |
| Open registry of sanitized metadata and hashes | An archive of raw payloads or private data |

Nearby tools answer different questions: model and data cards describe a system as
written by its authors, leaderboards rank systems, supply-chain provenance
(SLSA, in-toto) covers how software was built, and fact-check markup records a
publisher's verdict. ClaimBound records what one **frozen, reproducible check of a
public claim** found, and how far anyone has independently reproduced it.

## Public Workflows

| Workflow | Question it helps answer | Start here |
| --- | --- | --- |
| Public AI transparency | Did a public system-card or model-card **source boundary** pass a frozen audit? | [Current evidence tracks](https://github.com/ClaimBound/claimbound-evidence/blob/main/docs/CURRENT_EVIDENCE_TRACKS.md) |
| European and public open data | Did an official public-data source pass a source-boundary or narrow empirical gate? | [Evidence card examples](https://github.com/ClaimBound/claimbound-evidence/blob/main/docs/evidence_cards/README.md) |
| Software development evidence | Did a narrow build, validator or regression claim pass under fixed commands? | [Software development workflow](https://github.com/ClaimBound/claimbound-evidence/blob/main/docs/SOFTWARE_DEVELOPMENT_WORKFLOW.md) |

Blocked cards, artifact-only records and optional advanced protocol layers are
documented separately. They are not hidden failures. See
[future applications](https://github.com/ClaimBound/claimbound-evidence/blob/main/docs/future_applications/README.md),
[artifacts catalog](https://github.com/ClaimBound/claimbound-evidence/blob/main/docs/artifacts/README.md) and
[advanced optional docs](https://github.com/ClaimBound/claimbound-evidence/blob/main/docs/README.md#advanced-optional).

## Example Cards

Each SVG uses separate chips. **Green** and **red** on the result-status chip
mean a narrow gate pass or honest non-pass. **Yellow** appears on the
**reproduction** chip when `reproduction_level` records source-byte drift — not
on the gate outcome itself. None of this implies general model quality.

<p>
  <img
    src="https://raw.githubusercontent.com/ClaimBound/claimbound-evidence/main/docs/evidence_cards/CLAIMBOUND-ANTHROPIC_SYSTEM_CARDS_SOURCE_AUDIT_D001-2026-05-08.svg"
    alt="Green ClaimBound card for Anthropic system-card source audit"
    width="32%"
  />
  <img
    src="https://raw.githubusercontent.com/ClaimBound/claimbound-evidence/main/docs/evidence_cards/CLAIMBOUND-NASA-POWER-D103-2026-04-29.svg"
    alt="ClaimBound card with green gate pass and yellow reproduction chip for NASA POWER source-byte drift"
    width="32%"
  />
  <img
    src="https://raw.githubusercontent.com/ClaimBound/claimbound-evidence/main/docs/evidence_cards/CLAIMBOUND-NOAA-COOPS-D131-2026-04-30.svg"
    alt="Red ClaimBound card for NOAA negative result"
    width="32%"
  />
</p>

More examples, including EU source audits and blocked-source cards:
[evidence cards](https://github.com/ClaimBound/claimbound-evidence/blob/main/docs/evidence_cards/README.md) ·
[registry index](https://github.com/ClaimBound/claimbound-evidence/blob/main/docs/registry/evidence_index.json)

## For Reviewers And External Operators

- [Reviewer path](https://github.com/ClaimBound/claimbound-evidence/blob/main/docs/REVIEWER_PATH.md) — one-page orientation and credibility gaps
- [Reviewer summary](https://github.com/ClaimBound/claimbound-evidence/blob/main/docs/REVIEWER_SUMMARY.md) — problem, strongest cards, baseline
  versus roadmap, differentiation from adjacent projects
- [Common misreadings](https://github.com/ClaimBound/claimbound-evidence/blob/main/docs/COMMON_MISREADINGS.md) — green ≠ safe, yellow drift FAQ
- [European Dimension](https://github.com/ClaimBound/claimbound-evidence/blob/main/docs/EUROPEAN_DIMENSION.md) — EU open-data angle and limits
- [Volunteer one-pager](https://github.com/ClaimBound/claimbound-evidence/blob/main/docs/external_verification/VOLUNTEER_ONE_PAGER.md) — close
  VERIFY #85–#87 (~15 min first time)
- [External verification packs](https://github.com/ClaimBound/claimbound-evidence/blob/main/docs/external_verification/README.md) — split VERIFY
  work across Tier A/B/C operators
- [External operator starter pack](https://github.com/ClaimBound/claimbound-evidence/blob/main/docs/EXTERNAL_OPERATOR_STARTER_PACK.md) — read,
  rerun, drift report or boundary question paths
- [Planned work not shipped](https://github.com/ClaimBound/claimbound-evidence/blob/main/docs/PLANNED_NOT_SHIPPED.md) — roadmap vs current code
- [Artifacts catalog](https://github.com/ClaimBound/claimbound-evidence/blob/main/docs/artifacts/README.md) — NYC TLC / CDC artifact-only records
- [Public roadmap 2026](https://github.com/ClaimBound/claimbound-evidence/blob/main/docs/ROADMAP_2026.md) — planned open-source hardening work

Independent from [EviBound](https://arxiv.org/abs/2511.05524); see
[scope split](https://github.com/ClaimBound/claimbound-evidence/blob/main/docs/RELATED_WORK_AND_INDEPENDENCE.md). Not affiliated with
commercial or similarly named ClaimBound / Claimbound projects or domains
(including on GitHub or GitLab); see
[similar names](https://github.com/ClaimBound/claimbound-evidence/blob/main/docs/RELATED_WORK_AND_INDEPENDENCE.md#similar-names-and-commercial-lookalikes).
Canonical home: GitHub
[ClaimBound/claimbound-evidence](https://github.com/ClaimBound/claimbound-evidence);
the GitLab repo is a public mirror only.

## Install

Works on Windows, macOS and Linux. See [platform support](https://github.com/ClaimBound/claimbound-evidence/blob/main/docs/PLATFORM_SUPPORT.md).

From PyPI — card-level CLI (`validate-card`, `inspect card`, `hash`, `run-root`):

```bash
pipx install claimbound-evidence      # or: uv tool install claimbound-evidence / pip install claimbound-evidence
claimbound --version
claimbound validate-card path/to/card.json
```

`validate-all`, `demo`, `rerun`, `drift` and `verify` need the cards, registry and
scripts of a repository clone (see below); from a pip install they print how to get one.
The full guide, including a lighter sparse clone, is the
[documentation site](https://claimbound.github.io/claimbound-evidence/docs/).

From source (dev extras + full checkout):

```bash
git clone https://github.com/ClaimBound/claimbound-evidence.git
cd claimbound-evidence
uv sync --extra dev
uv run claimbound doctor
```

## Reproduce In Three Commands

```bash
uv run claimbound validate-all
uv run --extra dev python -m pytest -q
uv run claimbound demo eea-source-audit
```

If `validate-all` passes, the committed cards and registry match the current
validators. A scaffold or demo output is not evidence until a protocol is frozen,
the run completes, the card validates and the registry is updated.

## Quick Start

```bash
uv run claimbound new
uv run claimbound demo grok-source-audit
uv run claimbound run-root \
  --protocol-id EXAMPLE_D001 \
  --source-url https://example.org/source \
  --operator your-name-or-handle
```

Non-interactive scaffold, EU runners, card protocols and operator runbooks:
[getting started](https://github.com/ClaimBound/claimbound-evidence/blob/main/docs/GETTING_STARTED.md).

Portable AI control rules for evidence-bound work:
[12 AI Life Rules](https://github.com/ClaimBound/claimbound-evidence/blob/main/docs/TWELVE_AI_LIFE_CONTROLS.md).

## Boundary

This repository is an independently usable open evidence foreground. It does not
include or require private background technology.

Evidence cards are reusable examples and validation records, not a hosted review
service, legal advice, or on-demand third-party check queue. The registry stores
sanitized metadata and hashes, not raw payloads.

See [governance](https://github.com/ClaimBound/claimbound-evidence/blob/main/GOVERNANCE.md), [maintainer boundary](https://github.com/ClaimBound/claimbound-evidence/blob/main/MAINTAINER_BOUNDARY.md) and
[release process](https://github.com/ClaimBound/claimbound-evidence/blob/main/RELEASE_PROCESS.md).

## Community

- [Contributing guide](https://github.com/ClaimBound/claimbound-evidence/blob/main/CONTRIBUTING.md)
- [Code of Conduct](https://github.com/ClaimBound/claimbound-evidence/blob/main/CODE_OF_CONDUCT.md)
- [Security policy](https://github.com/ClaimBound/claimbound-evidence/blob/main/SECURITY.md)
- [Discussions](https://github.com/ClaimBound/claimbound-evidence/discussions)

## License

Apache-2.0. See [LICENSE](https://github.com/ClaimBound/claimbound-evidence/blob/main/LICENSE).
