# arseniygrach3419/cat-breeds-dirty-dataset

- Platform link: https://sourcecraft.dev/arseniygrach3419/cat-breeds-dirty-dataset
- Snapshot date: 2026-10-08
- Generated at: 2026-10-09 09:00 UTC

## Score

**31–45** (F–D — starting point–room to grow), applicable weight 55%.

## Categories

| Category | Weight (nominal / applied) | State | Score | Coverage |
| --- | --- | --- | --- | --- |
| Documentation and best practices | 20 / 36.36 | scored | 31.75 | 100% |
| CI/CD | 15 / 27.27 | partial | 0–50 | 50% |
| Security | 20 / 0 | no data | – | 0% |
| Project activity | 20 / 36.36 | scored | 54.54 | 100% |
| Issues | 15 / 0 | no data | – | 0% |
| Code health | 10 / 0 | no data | – | 0% |

## Strengths

- **README** (Documentation and best practices): earned 10.39 of 10.39
- **last commit** (Project activity): earned 6.06 of 6.06

## Recommendations

- **add a SourceCraft CI configuration (.sourcecraft/ci.yaml); the service does not see external CI** [critical] (CI/CD) — SourceCraft CI configuration, may return up to +13.64 (the build and the checks are reproducible for whoever takes the code)
- **publish a release or tag a version** [critical] (Project activity) — last release, may return up to +6.06 (the build and the checks are reproducible for whoever takes the code)
- **add a license file** [medium] (Documentation and best practices) — license file, may return up to +3.46 (the terms of use are stated, not guessed)
- **write tests for the project** [medium] (Documentation and best practices) — tests directory, may return up to +3.46 (the build and the checks are reproducible for whoever takes the code)
- **write project documentation** [medium] (Documentation and best practices) — docs directory, may return up to +3.46 (a newcomer can tell how to run the project and where to look)
- **containerize the project — add a Dockerfile** [medium] (Documentation and best practices) — Dockerfile, may return up to +3.46 (a newcomer can tell how to run the project and where to look)
- **add a dependency lockfile** [medium] (Documentation and best practices) — dependency lockfile, may return up to +3.46 (the build and the checks are reproducible for whoever takes the code)
- **set up a review policy** [medium] (Documentation and best practices) — review policy, may return up to +3.46 (a change is reviewed, and the history stays readable)
- **commit more often** [medium] (Project activity) — commits in the last year, may return up to +3.03 (the repository shows it is alive, not abandoned)
- **bring in a second maintainer** [medium] (Project activity) — unique authors, may return up to +3.03 (the work does not rest on a single person)
- **spread work across more weeks instead of bursts** [medium] (Project activity) — weekly activity, may return up to +2.73 (the repository shows it is alive, not abandoned)
- **add a contributing guide** [optional] (Documentation and best practices) — contributing guide, may return up to +1.73 (an outside contributor gets an answer, and their work is not lost)
- **set up a branch protection policy** [optional] (Documentation and best practices) — branch protection policy, may return up to +1.73 (a change is reviewed, and the history stays readable)
- **share the work with other contributors** [optional] (Project activity) — top author's share of commits, may return up to +1.68 (the work does not rest on a single person)
- **write meaningful commit messages** [optional] (Documentation and best practices) — share of short commit messages, may return up to +0.58 (a change is reviewed, and the history stays readable)

## Not measured — interval width

Not a penalty: the lower bound of the score counts the metric as zero, the upper bound as full.

- **CI/CD** — partial
  - CI runs: up to −13.64 — access to the source is closed

## Not counted in the score, and why

- **Security** — no data
  - the basis was not saved: there was nothing to measure, or the reason was not recorded in the snapshot
- **Issues** — no data
  - the basis was not saved: there was nothing to measure, or the reason was not recorded in the snapshot
- **Code health** — no data
  - the basis was not saved: there was nothing to measure, or the reason was not recorded in the snapshot