L’interface est localisée ; les présentations et documents des Skills sont traduits automatiquement par IA dans votre langue. Le code, les commandes, chemins et termes techniques restent en version originale, et le document de l’auteur fait foi.
Explorer les Skills
Données et recherche · SKILL.md

data-and-storage

data-and-storage

Manage Galaxy datatypes, metadata, dataset storage, object stores, file sources, data managers, tool data tables, and storage safety boundaries.

Independent GitHub snapshot · Automated classification · Original documentation governs

Documentation de la Skill

GitHub

name: data-and-storage description: "Manage Galaxy datatypes, metadata, dataset storage, object stores, file sources, data managers, tool data tables, and storage safety boundaries." disable-model-invocation: true metadata: disco-role: operating license: NOASSERTION

Galaxy Data and Storage

Use this sub-skill when a task involves Galaxy datatypes, sniffers, datatype metadata, dataset storage, object stores, file sources, data managers, tool data tables, .loc entries, or storage cleanup risk assessment.

Start Here

  • For datatype implementation, registration, metadata, composite formats, and sniffer safety, read Data Formats.
  • For data manager XML/JSON contracts and tool-data table entries, read Data Managers.
  • For object-store, file-source, user template, credential, and storage-test planning, read Object Stores and Files.
  • For common failures and conservative triage, read Troubleshooting.
  • For a quick bounded-sniffer review, run python scripts/check_datatype_sniffer.py PATH_OR_STDIN.

Routing Boundaries

  • Stay here for datatype classes, datatypes_conf.xml registration, sniff, set_meta, set_peek, data-manager outputs, tool-data tables, object-store configuration concepts, file-source credentials, dataset cleanup safety, and data/storage tests.
  • Route broad Galaxy config placement, startup, galaxy.yml editing, and admin deployment planning to Configuration and Admin.
  • Route tool XML mechanics, wrapper tests, workflow YAML, and dependency declarations to Workflows and Tools.
  • Route upload/download API calls, history dataset automation, and API-client scripting to API Automation.

Safety Defaults

  • Never run cleanup, purge, migration, cloud object-store, or credentialed file-source actions without explicit user approval, target confirmation, backup/rollback context, and dry-run or read-only checks first.
  • Treat sniffers as production hot paths: they must inspect bounded prefixes or bounded header/sample helpers, never entire files.
  • Treat user file sources and object-store templates as secret-bearing configuration; do not print, persist, or invent credentials.
  • Prefer local/unit tests and template/model validation before remote integration tests; skip cloud or network tests unless credentials and disposable resources are explicitly supplied.
Skills enregistrées