Marketplace

171 skills

Each listing spells out exactly what's inside β€” the steps, the ready-to-run code, and which AI tools it works with β€” so you know what you're getting before you buy.

Clear filters

Use when creating, updating, suspending or deleting Google Workspace users, groups, org units or licences at scale via the Admin SDK Directory API.

  • βœ“11-step procedure
  • βœ“Runnable Python included
  • βœ“5-point quality checklist
cloudBest with a strong model (claude-sonnet-4-5)Β·8 min readΒ·Updated 1 month ago

Use when auditing a Google Workspace tenant: admin roles, 2SV coverage, third-party OAuth grants, external Drive sharing and login anomalies.

  • βœ“7-step procedure
  • βœ“Runnable Python included
  • βœ“6-point quality checklist
cloudBest with a strong model (claude-opus-5)Β·8 min readΒ·Updated 1 month ago

Use when a task needs domain-wide Google Workspace API access (Directory, Reports, Gmail, Drive) and credentials are missing or returning unauthorized_client.

  • βœ“12-step procedure
  • βœ“Runnable Python / Shell / JSON included
  • βœ“8-point quality checklist
cloudBest with a strong model (claude-opus-4-6)Β·9 min readΒ·Updated 1 month ago

Diagnose visual bugs in Godot 4 β€” invisible objects, blurry sprites, black materials, transparency sorting, shadow and shader faults β€” using measured evidence.

  • βœ“4-step procedure
  • βœ“Runnable GDScript / Shell included
  • βœ“9-point quality checklist
code-qualityBest with a strong model (Claude Opus 5)Β·15 min readΒ·Updated 1 month ago

Use when building or hardening a GCP organization: folder/project hierarchy, Shared VPC, org policies, least-privilege IAM, log sinks, billing guardrails.

  • βœ“10-step procedure
  • βœ“Runnable Shell / Terraform included
  • βœ“10-point quality checklist
cloudBest with a strong model (Claude Opus 5)Β·12 min readΒ·Updated 1 month ago

Design and implement ML feature stores: online/offline serving, feature freshness, training-serving skew prevention, and feature discovery.

  • βœ“9-step procedure
  • βœ“6 pitfalls to avoid
  • βœ“Installs into 6 tools
dataBest with a strong model (Claude Sonnet 4)Β·3 min readΒ·Updated 1 month ago

Design ML experiment tracking: hyperparameter logging, model versioning, reproducibility, and experiment comparison workflows.

  • βœ“9-step procedure
  • βœ“6 pitfalls to avoid
  • βœ“Installs into 6 tools
dataBest with a strong model (Claude Sonnet 4)Β·3 min readΒ·Updated 1 month ago

Design statistically rigorous A/B tests: hypothesis formation, sample size calculation, metric selection, and guardrail setup.

  • βœ“8-step procedure
  • βœ“7 pitfalls to avoid
  • βœ“Installs into 6 tools
productWorks with any modern AI assistantΒ·3 min readΒ·Updated 1 month ago

Design event-driven systems: event schema design, broker selection, idempotency, ordering guarantees, and event sourcing.

  • βœ“10-step procedure
  • βœ“7 pitfalls to avoid
  • βœ“Installs into 6 tools
code-qualityWorks with any modern AI assistantΒ·4 min readΒ·Updated 1 month ago

Invoke when an AI system touches the EU market and you must classify its EU AI Act risk tier and output the obligations and article citations that follow.

  • βœ“11-step procedure
  • βœ“Runnable Python included
  • βœ“8-point quality checklist
industry-verticalBest with a strong model (Claude Opus 5)Β·11 min readΒ·Updated 1 month ago

Design and scaffold idempotent, observable extract-transform-load pipelines with safe retries, backfills, and quality gates; use when building data pipelines.

  • βœ“9-step procedure
  • βœ“2 ready-to-run code blocks
  • βœ“8-point quality checklist
dataBest with a strong model (Claude Sonnet 4)Β·6 min readΒ·Updated 1 month ago

Audit error handling across a codebase: unhandled rejections, swallowed exceptions, missing retries, and inconsistent error types.

  • βœ“8-step procedure
  • βœ“6 pitfalls to avoid
  • βœ“Installs into 6 tools
code-qualityWorks with any modern AI assistantΒ·3 min readΒ·Updated 1 month ago

Use before running untrusted or model-generated code, installs, or binaries: stand up a disposable, network-restricted sandbox, run it, then tear it down.

  • βœ“8-step procedure
  • βœ“Runnable Shell included
  • βœ“9-point quality checklist
devopsWorks with any modern AI assistantΒ·7 min readΒ·Updated 1 month ago

Use when onboarding, transferring, disabling or offboarding Microsoft Entra ID users, their group memberships and their licenses through Microsoft Graph.

  • βœ“10-step procedure
  • βœ“Runnable Python / JSON included
  • βœ“9-point quality checklist
cloudBest with a strong model (claude-opus-4-6)Β·11 min readΒ·Updated 1 month ago

Use when designing, staging or rolling out Microsoft Entra Conditional Access policies without locking the tenant out.

  • βœ“10-step procedure
  • βœ“Runnable Python included
  • βœ“9-point quality checklist
cloudNeeds a top-tier modelΒ·11 min readΒ·Updated 1 month ago

Use when a task needs Microsoft Graph access: register an Entra app, pick delegated vs application permissions, grant admin consent, authenticate.

  • βœ“10-step procedure
  • βœ“Runnable Python / JSON included
  • βœ“7-point quality checklist
cloudBest with a strong model (claude-opus-5)Β·11 min readΒ·Updated 1 month ago

Choose embedding models and strategies: dimensionality, domain-specific models, multi-modal embeddings, and evaluation.

  • βœ“9-step procedure
  • βœ“6 pitfalls to avoid
  • βœ“Installs into 6 tools
dataBest with a strong model (Claude Sonnet 4)Β·3 min readΒ·Updated 1 month ago

Design document processing pipelines: OCR, layout analysis, information extraction, classification, and human review.

  • βœ“8-step procedure
  • βœ“6 pitfalls to avoid
  • βœ“Installs into 6 tools
dataBest with a strong model (Claude Sonnet 4)Β·4 min readΒ·Updated 1 month ago

Use when a Dockerfile builds slowly, produces oversized images, or needs hardening; rewrites it for smaller, faster, and more secure builds.

  • βœ“10-step procedure
  • βœ“1 ready-to-run code block
  • βœ“11-point quality checklist
devopsBest with a strong model (Claude Sonnet 4)Β·6 min readΒ·Updated 1 month ago

Design disaster recovery strategies: RTO/RPO targets, backup automation, failover procedures, and DR testing.

  • βœ“10-step procedure
  • βœ“6 pitfalls to avoid
  • βœ“Installs into 6 tools
devopsBest with a strong model (Claude Sonnet 4)Β·5 min readΒ·Updated 1 month ago

Deploy and manage DigitalOcean infrastructure: Droplets, App Platform, Kubernetes, managed databases, spaces, and networking via doctl CLI.

  • βœ“Runnable Shell included
  • βœ“7 pitfalls to avoid
  • βœ“Installs into 6 tools
devopsBest with a strong model (Claude Sonnet 4)Β·6 min readΒ·Updated 1 month ago

Use when deploying or operating apps on DigitalOcean App Platform and Managed Databases: app specs, secrets, PRE_DEPLOY migrations, domains, scaling, cost.

  • βœ“14-step procedure
  • βœ“Runnable YAML / Terraform / Shell included
  • βœ“7-point quality checklist
cloudWorks with any modern AI assistantΒ·12 min readΒ·Updated 1 month ago

Design and implement design systems: token architecture, component APIs, documentation patterns, and governance.

  • βœ“7-step procedure
  • βœ“5 pitfalls to avoid
  • βœ“Installs into 6 tools
frontendWorks with any modern AI assistantΒ·3 min readΒ·Updated 1 month ago

Manage API and feature deprecations safely: sunset timelines, migration guides, usage tracking, and forced upgrade paths.

  • βœ“9-step procedure
  • βœ“7 pitfalls to avoid
  • βœ“Installs into 6 tools
code-qualityWorks with any modern AI assistantΒ·4 min readΒ·Updated 1 month ago