Dataset Viewer
The dataset could not be loaded because the splits use different data file formats, which is not supported. Read more about the splits configuration. Click for more details.
Couldn't infer the same data file format for all splits. Got {NamedSplit('validation'): (None, {}), NamedSplit('test'): ('json', {})}
Error code:   FileFormatMismatchBetweenSplitsError

Need help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.

AriaImageGuard container hardening lab

Synthetic Dockerfile, Compose, and pre-generated inspect JSON for the AriaImageGuard configuration-review demo: a privileged shop web, a hardened lookalike, socket/capability escape, a missing-USER case that stays unknown, and a secret/ADD/curl|sh build.

Organization dataset and collection are public. Live Gradio is alirezaaminzadeh/aria-image-guard and the organization card is AriaAICompany/aria-image-guard. Runnable Space source is stored in demo/. Collection: Aria AI — Cybersecurity.

Generated data, seed 18. Image names, tokens, and hosts are laboratory placeholders. The control catalog is hadolint-cis-docker-mapped-v1 (2026-09-20). Hadolint and Docker are not executed on this dataset. A missing USER is not a complete runtime verdict. Package CVEs belong to AriaSBOM.

Files

  • data/scenarios.jsonl
  • data/expected.json
  • data/schema.json
  • data/splits.json
  • data/controls.json
  • data/evaluation.json
  • data/sample_preview.json
  • data/stacks/ — five scenario folders
  • demo/ — runnable Gradio Space source
  • showcase/ — organization static card

Split

Development vs test is by scenario family: privileged runtime and socket/caps vs lookalike, missing USER, and build hygiene. Renaming one service of the same finding is not placed across the split.

License

CC-BY-4.0. Keep the synthetic-data and laboratory-config labels.

Downloads last month
51

Collection including AriaAICompany/image-guard-lab