Google Cloud Sustainability Advisor
Evaluates Google Cloud workloads against the WAF Sustainability pillar and recommends practical emissions-reduction actions.
The skill provides guidance, questions, and checklists only; it does not execute commands, access credentials, upload data, or apply destructive changes, so privilege and external-effect risk is low. However, it lacks explicit sensitive-data handling, user confirmation, data-flow disclosure, rollback, and audit guidance, and its listed source links are not accompanied by a verification procedure, so points are deducted.
The document is internally coherent and supplies principles, product examples, assessment questions, and a validation checklist, making the happy path understandable. It provides no input/output contract, abnormal-input handling, failure feedback, compatibility guidance, or test suite, and product freshness cannot be established statically; the score is therefore kept within the static-review limit and reduced.
The intended scenario is clear: assessing Google Cloud workload sustainability. Coverage spans architecture, regions, resources, software, storage, measurement, and culture, but triggers, non-fit boundaries, required inputs, and output format are unspecified, with no Chinese-language support guidance. The core function is not wholly dependent on external services, although referenced documentation may be difficult to reach from mainland China, so points are deducted.
The skill has readable layered sections; the README documents installation and indexing; the repository supplies an Apache-2.0 license, an issue path, and an active-development note. It lacks skill-specific dependency notes, example outputs, FAQs, known limitations, versioning, changelog, and clearly assigned maintenance responsibility, so points are deducted.
The principles, product mapping, assessment questions, and checklist can directly support a manual sustainability review, so the core value is evident. However, there is no fixed output format, representative example, quantitative method, prioritization model, or decision boundary; results still require substantial human validation, so only a moderate static-review score is justified.
Most core principles include links to Google Cloud documentation, providing some primary-source traceability. There is no committed test suite, CI coverage, execution record, independent review, or versioned corroboration, and no reproduction was performed in this static review, so evidence coverage remains limited and points are deducted.
- Do not treat product capabilities, regional carbon indicators, or storage-energy claims as currently verified facts; check each against the relevant official documentation before use.
- The skill defines no fixed output format, prioritization method, or risk boundary, so generated recommendations require review by cloud architecture, finance, and sustainability owners.
- Referenced Google Cloud documentation may have inconsistent reachability from mainland-China network environments.
What does this skill do, and when should you use it?
This skill provides sustainability-focused architecture guidance for Google Cloud workloads. It covers low-carbon regions, AI/ML energy efficiency, resource utilization, software efficiency, data and storage, carbon measurement, governance, and organizational culture. It asks workload-specific questions and applies a validation checklist to assess alignment with the recommendations. It fits teams that want to include environmental impact in Google Cloud architecture, operations, or GreenOps reviews.
Generates guidance based on the Sustainability pillar of the Google Cloud Well-Architected Framework; asks questions about region carbon intensity, scaling, AI/ML efficiency, software and end-user efficiency, data storage, carbon analysis, and governance; gives recommendations involving example products and features such as Carbon Footprint, BigQuery, Region Picker, Active Assist, Cloud Run, GKE Autopilot, Cloud Batch, Spot VMs, Cloud Storage Lifecycle Management, and Cloud TPUs; and evaluates architectures with a topic-based checklist covering regional selection, resource optimization, software efficiency, storage governance, carbon measurement, and sustainability culture.
- A cloud architect designing a new Google Cloud workload needs to balance carbon intensity, cost, and latency when selecting regions.
- A platform engineering team wants to reduce idle compute through autoscaling, rightsizing, managed services, or Spot VMs.
- An AI/ML team is assessing TPU usage, model compression, or PEFT to reduce computational complexity and improve computations per watt.
- An application team is reviewing polling, caching, asset size, and end-user processing to identify avoidable energy use.
- A FinOps or GreenOps team needs to analyze Carbon Footprint data with BigQuery to identify emissions hotspots.
- An enterprise governance team wants to connect sustainability targets, carbon metrics, training, and industry guidance to technical decisions.
What are this skill's strengths and limitations?
- Covers sustainability from region selection and resource use through software, storage, measurement, governance, and culture.
- Includes concrete assessment questions and a checklist suitable for architecture reviews.
- Names relevant Google Cloud products and features, including Carbon Footprint, BigQuery, Region Picker, and Active Assist.
- The repository is licensed under Apache-2.0, and its README describes the project as under active development.
- It is centered on the Google Cloud WAF Sustainability pillar rather than being a general multi-cloud sustainability evaluator.
- The source provides no scripts, automated execution, test suite, or quantitative scoring system.
- The source does not document client-specific invocation protocols, output formats, or required permissions.
- Recommendations depend on workload context and carbon data; the skill does not claim to retrieve that data automatically.
How do you install this skill?
Run npx skills add google/skills and select the specific skill during installation. The target skill is skills/cloud/google-cloud-waf-sustainability/SKILL.md. The source does not document a more specific client directory or version requirement.
How do you use this skill?
After installing it in an Agent Skills-compatible client, submit a request such as: Assess this Google Cloud workload for sustainability risks, ask the necessary questions first, then provide prioritized recommendations using the validation checklist. The source does not define a specific command, parameter set, or invocation syntax.