Skip to main content

research

Before designing a feature, research how comparable libraries solve the problem, check best practices, known pitfalls, and upstream changes. Use when starting a new feature, investigating an unfamiliar area, or the user asks to research something.

Informations de source

Dépôt
NVIDIA/multi-storage-client
Dernière activité de la source
9 septembre 2026 à 18:30
Langue détectée de SKILL.md
anglais
Étoiles
87
Forks
28

Options d'installation

Le prompt qui vérifie d'abord la source est sélectionné par défaut. Vous pouvez passer à une commande directe ou télécharger une copie locale.

Vérifiez les fichiers source

Lisez SKILL.md et les fichiers associés affichés par SkillsMP avant de décider de l'installer.

Affichage de SKILL.md

SKILL.md
Instructions source · Aperçu en lecture seule
name
research
description
Before designing a feature, research how comparable libraries solve the problem, check best practices, known pitfalls, and upstream changes. Use when starting a new feature, investigating an unfamiliar area, or the user asks to research something.
invocable
auto
# Research Research the problem space before designing or implementing a feature to avoid reinventing the wheel and catch issues early. ## When to Use - Starting a new feature or significant enhancement. - Working with unfamiliar APIs, cloud SDKs, or library internals. - User explicitly asks to research or investigate an approach. ## Workflow ### 1. Clarify scope Confirm with the user: - What problem are we solving? - Which language layers are involved (Python, Rust)? - Any known constraints or preferences? ### 2. Research comparable solutions Search for: - How other multi-cloud storage libraries solve the same problem. - Best practices for the relevant cloud SDKs (AWS S3, GCS, Azure, OCI). - Known pitfalls or breaking changes in upstream dependencies. - PyO3/maturin patterns (if Rust bindings are involved). ### 3. Check upstream Look for: - Recent releases of relevant dependencies. - Open issues or PRs related to the feature area. - Deprecation notices. ### 4. Produce a summary Output using this format: ```markdown ## Research Summary: <topic> ### Approaches Found - <approach 1>: <trade-offs> - <approach 2>: <trade-offs> ### Recommendation - <recommended approach with rationale> ### Risks / Blockers - <anything that could derail implementation> ### References - <links to docs, issues, examples> ```
Voir sur GitHub