[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6996BXYMW9ACK50KZR95S0E":3,"skill-same:01M6996BXYMW9ACK50KZR95S0E":111},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":26,"publishedAt":48,"linkOnly":49,"markdownContent":50,"markdownBody":51,"frontMatter":52,"refName":56,"commitSha":57,"skillDirectory":58,"forksCount":59,"repositoryHtmlUrl":60,"repositoryHomepageUrl":6,"ownerHtmlUrl":61,"ownerBlogUrl":62,"ownerDescription":63,"updatedAt":64,"install":65,"resources":78},"01M6996BXYMW9ACK50KZR95S0E","google-cloud-solution-build-deploy-agents",null,"Designs, builds, and deploys AI agents or multi-agent systems on Google Cloud. Provides an interactive workflow to gather requirements, recommend a tailored architecture, and generate deployment instructions. Use when designing or implementing agentic systems on Google Cloud. Don't use for general Google Cloud solution architecture (use google-cloud-solution-architecture instead) or for narrow tasks targeting a single product without agent context.","google\u002Fskills","google","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F1342004?v=4","01M698VYEFDM0VJBZDYYJ2T83S","Google","01M698WR0H76ENHE587T9EJT89",24122,"Apache-2.0","https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fcloud\u002Fgoogle-cloud-solution-build-deploy-agents\u002FSKILL.md",[18,23],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000011","运维与部署","SCENARIO","使用场景",{"categoryNo":24,"categoryName":25,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000023","Skill与智能体开发",[27,28,9,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47],"google cloud solution build deploy agents","google cloud","solution build","build","cloud","deploy","tasks","requirements","workflow","when","use","targeting","single","general","implementing","gather","deployment","for","ai agents","agentic","google cloud solution architecture","2026-11-10 10:12:30",false,"---\nname: google-cloud-solution-build-deploy-agents\nmetadata:\n  version: \"1.0.2\"\n  category: MultiProductSolutions\ndescription: >-\n  Designs, builds, and deploys AI agents or multi-agent systems on Google Cloud.\n  Provides an interactive workflow to gather requirements, recommend a tailored\n  architecture, and generate deployment instructions. Use when designing or\n  implementing agentic systems on Google Cloud. Don't use for general Google\n  Cloud solution architecture (use google-cloud-solution-architecture instead)\n  or for narrow tasks targeting a single product without agent context.\n---\n\n# Build and deploy AI agents on Google Cloud\n\nThis skill guides agents through the workflow of designing and implementing a\ntailored multi-product solution in the cloud for a given workload, use case, or\nrequirement.\n\n## Attribution\n\n**MUST READ**: Before executing any `gcloud`, REST API, or MCP command, you\nMUST read and follow the instructions in\n[Usage Attribution](references\u002Fattribution.md).\n\n## Workflow\n\nThe solution design and implementation workflow is divided into the following\nphases:\n\n*   **Phase 1: Requirements discovery and analysis**: Analyze the workload's\n    requirements, constraints, dependencies, and current state.\n*   **Phase 2: Solution design**: Build a technology stack, architecture, and\n    deployment configuration for the workload based on Google Cloud design best\n    practices and recommendations.\n*   **Phase 3: Implementation plan**: Generate automation and instructions to\n    deploy the solution.\n*   **Phase 4: Solution validation**: Validate that the deployment meets the\n    requirements of the workload.\n\nCopy this checklist into your active task\u002Fplan artifact to track progress across\nthe four phases:\n\n-   [ ] Phase 1: Requirements discovery and analysis completed & confirmed.\n-   [ ] Phase 2: Solution architecture generated & approved.\n-   [ ] Phase 3: Implementation plan generated & approved.\n-   [ ] Phase 4: Solution validation generated & approved.\n\n### Phase 1: Requirements discovery and analysis\n\n1.  **Discover requirements**: Gather and understand the functional and\n    non-functional requirements, business goals, and current state (if any) of\n    the workload, including its architecture, dependencies, and constraints.\n\n    *Important*: First, check whether the user's initial prompt has already\n    answered the following questions or whether the prompt explicitly asks you\n    to propose a solution architecture\u002Fdiagram from a given set of parameters.\n\n    -   If the user's prompt provides sufficient requirements and it explicitly\n        requests an architecture proposal or diagram, then skip asking the\n        questions below, and instead proceed to the step **Recommend agent\n        design pattern**.\n    -   If the user's prompt doesn't provide sufficient requirements, then\n        complete these steps to gather missing information:\n\n        1.  Ask the user to describe the functional requirements of their\n            workload: business processes, activities, and use cases.\n\n        2.  Ask the user to describe the non-functional requirements (security,\n            privacy, compliance, reliability, disaster recovery, cost,\n            operations, performance, and sustainability) of their workloads.\n\n        3.  Ask the user what existing systems, knowledge bases, product\n            documentation, or other documentation the AI agents need to access\n            for grounded guidance.\n\n        4.  Ask the user to describe dependencies, if any, on other workloads,\n            products, or tools.\n\n        5.  Review the input that the user has provided so far, and check\n            whether there are any ambiguities or contradictions in the input.\n\n            If you identify any ambiguities or contradictions in the\n            requirements that the user has provided, then do the following for\n            each ambiguity or contradiction that you identify:\n\n            *   Describe the ambiguity or contradiction.\n            *   Ask the user how they wish to resolve the ambiguity or\n                contradiction.\n                *   If the user delegates the choice to you (e.g., the user\n                    replies with \"do what you think is best\" or \"you decide\"),\n                    then provide a clear suggestion to resolve the ambiguity or\n                    contradiction, explain your reasoning, and ask the user to\n                    approve your suggestion.\n\n            **Critical**: Until all the ambiguities and contradictions that you\n            identify are resolved according to the preceding guidance, you must\n            NOT recommend or generate any architecture design, technical\n            decomposition, or Google Cloud product recommendations.\n\n2.  **Recommend agent design pattern**: Evaluate the complexity, workflow,\n    latency, and cost requirements of the workload to recommend an agent design\n    pattern:\n\n    -   **Single-agent system**: Recommend for simpler tasks, acting as an\n        effective starting point to refine core logic and tools.\n    -   **Multi-agent system**: Recommend for complex problems requiring\n        multiple specialized agents to collaborate on a workflow.\n\n3.  **Identify components**: Based on the requirements analysis, generate a\n    technical decomposition of the workload. The technical decomposition must\n    identify the logical components of the workloads and their relationships.\n    Also identify any cross-cloud components, hybrid components, or on-premises\n    components that the solution needs to integrate with.\n\n4.  **Ask for confirmation**: Ask the user to confirm whether the recommended\n    design pattern and technical decomposition match their workload\n    requirements.\n\n5.  **Iterate**: If the user requests changes, generate an updated technical\n    decomposition, and ask the user to confirm the changes. Continue iterating\n    until the user confirms the technical decomposition. Proceed to the next\n    phase only after the user provides confirmation of the technical\n    decomposition.\n\n### Phase 2: Solution design\n\n1.  **Retrieve relevant Google Cloud guidance from\n    `references\u002Frelated-guidance.md`**.\n\n    *Important*: Use the content that you retrieved from\n    `references\u002Frelated-guidance.md` to ground the guidance that you generate in\n    the remaining steps of this phase.\n\n2.  **Map components to Google Cloud products**: For each component in the\n    confirmed technical decomposition, identify the appropriate Google Cloud\n    products and features by consulting\n    [product-mappings.md](references\u002Fproduct-mappings.md) for detailed\n    recommendations, trade-offs, and alternatives across networking, frontends,\n    agent\u002Fmodel runtimes, memory stores, and tools.\n\n3.  **Create architecture diagram**: Create an architecture diagram in Mermaid\n    format: https:\u002F\u002Fgithub.com\u002Fmermaid-js\u002Fmermaid. The diagram should show the\n    components, their relationships, and data\u002Fcontrol flows.\n\n4.  **Generate design recommendations**: Generate design guidance based on the\n    following Google Cloud best practices and recommendations. Use the\n    information in `references\u002Frelated-guidance.md`, with an emphasis on the\n    guidance in `references\u002Fdesign-principles.md`.\n\n5.  **Draft solution architecture**: Compile the requirements, technical\n    decomposition, product mapping, architecture diagram, and design\n    recommendations into a single Markdown file adhering to the format in\n    [solution-template.md](assets\u002Fsolution-template.md). Save this document in\n    the workspace as `solution-architecture.md`.\n\n6.  **Request review**: Present the generated solution architecture (including\n    the complete fenced `mermaid` code block for the diagram) directly to the\n    user in your response, and explicitly request their feedback or approval.\n    When you present the architecture, ask the user to provide approval for you\n    to proceed with an implementation plan.\n\n7.  **Iterate**: If the user requests changes, generate an updated solution\n    architecture and repeat the steps from \"Map components to Google Cloud\n    products\" through \"Request review\" until the user approves the solution\n    architecture.\n\n### Phase 3: Implementation plan\n\n1.  **Retrieve relevant implementation resources**:\n\n    *Important*: Use the resources in references\u002Frelated-guidance.md as the\n    technical foundation for the Infrastructure as Code (IaC) and the deployment\n    instructions that you generate in the remaining steps of this phase.\n\n2.  **Identify deployment prerequisites**: Document prerequisites for the\n    deployment, including the following:\n\n    -   Projects and billing associations\n    -   Required Google Cloud APIs\n    -   Required IAM permissions\n    -   Any other prerequisites\n\n3.  **Generate Infrastructure as Code (IaC)**: Generate code (e.g., Terraform)\n    and deployment scripts to automate the provisioning of the proposed Google\n    Cloud resources.\n\n    -   Where appropriate, alongside or instead of raw infrastructure scripts,\n        instruct the user to set up or enhance the project structure, deployment\n        configuration, and CI\u002FCD pipelines by using Agents CLI commands:\n        -   To create a new project, use `agents-cli create`. Don't use\n            `agents-cli scaffold create`.\n        -   To add deployment and CI\u002FCD scaffolding to an existing repository,\n            use `agents-cli scaffold enhance`.\n\n4.  **Write deployment instructions**: Draft sequential, step-by-step deployment\n    instructions to execute the IaC and initialize the workload components.\n    Compile the deployment prerequisites, IaC, and deployment instructions into\n    a single Markdown file adhering to the format in\n    [implementation-template.md](assets\u002Fimplementation-template.md). Save this\n    document in the workspace as `implementation-instructions.md`.\n\n    -   The instructions MUST provide the exact ADK code to define a stateful\n        agent node that takes a prompt, calls a model, and returns a tool\n        execution request.\n    -   The instructions MUST demonstrate how to register tools like database\n        readers by using Model Context Protocol (MCP) standards.\n    -   If deploying the agent to Cloud Run, the instructions MUST show how to\n        configure Cloud Run to scale to zero when the agent is idle, reducing\n        runtime costs.\n    -   The instructions MUST recommend using encrypted environment variables to\n        store model parameters or private API credentials. Encryption helps to\n        prevent the exposure of sensitive credentials in plain-text container\n        log streams.\n    -   Where appropriate, the instructions MUST specify using the Agents CLI\n        `agents-cli deploy` command (alongside or instead of raw\n        infrastructure\u002Fdeployment scripts) to run the deployment.\n\n5.  **Request review**: Present the generated deployment instructions to the\n    user and explicitly request their feedback and confirmation.\n\n6.  **Iterate**: If the user requests changes, generate an updated\n    implementation plan and repeat the steps from \"Generate Infrastructure as\n    Code (IaC)\" through \"Request review\" until the user approves the\n    implementation plan.\n\n### Phase 4: Solution validation\n\n1.  **Retrieve relevant verification resources**:\n\n    *Important*: Use the resources in references\u002Frelated-guidance.md and their\n    verification patterns as the starting point for the validation checks and\n    verification scripts that you generate in the remaining steps of this phase.\n\n2.  **Define validation checks**: Outline validation steps to verify that the\n    deployed infrastructure meets the workload's requirements:\n\n    -   **Deployment dry-run**: Commands like `terraform plan` to preview\n        changes. Include instructions to run agent deployment in dry-run mode\n        (e.g., using `agents-cli deploy --dry-run` or `-n`) to preview steps and\n        Terraform executions before pushing to production.\n    -   **Local testing and quality verification**: Recommend using the Agents\n        CLI to run and test agent logic locally (`agents-cli run`) and conduct\n        systematic evaluations (`agents-cli eval run`) to verify agent quality\n        and performance before deploying.\n    -   **Connectivity and routing**: Verification of network paths, load\n        balancer routing, and service endpoints.\n    -   **Security policies**: Verification of restricted access, firewall\n        rules, and IAM enforcement.\n\n3.  **Generate verification scripts**: Draft lightweight scripts or command-line\n    instructions (e.g. using `curl`, `gcloud`, or `agents-cli`) that the user\n    can run to perform these validation checks.\n\n    -   The validation plan MUST include instructions using the Agents CLI for\n        local runs, evaluations, and post-deployment validation checks (e.g.,\n        `agents-cli run --url \u003Cservice-url>` to test the deployed service\n        endpoint).\n\n4.  **Compile validation plan**: Document the validation steps, verification\n    scripts, and expected outcomes in a single Markdown file adhering to the\n    format in [validation-template.md](assets\u002Fvalidation-template.md). Save this\n    document in the workspace as `validation-plan.md`.\n\n5.  **Request review**: Present the validation plan to the user and explicitly\n    request their feedback or approval on the validation plan.\n\n6.  **Conduct validation and finalize**: Assist the user in executing the\n    validation checks and troubleshooting any deployment issues. After the\n    solution is validated successfully, request final approval from the user.\n\n7.  **Iterate**: If the user requests changes, generate an updated validation\n    plan and repeat the steps from \"Define validation checks\" through \"Request\n    review\" until the user approves the validation plan.\n\n--------------------------------------------------------------------------------\n\n## References & Supporting Links\n\n*   For the complete list of Google Cloud architectural documentation, product\n    manuals, development kits, and checklists used by this skill, see\n    [related-guidance.md](references\u002Frelated-guidance.md).\n","# Build and deploy AI agents on Google Cloud\n\nThis skill guides agents through the workflow of designing and implementing a\ntailored multi-product solution in the cloud for a given workload, use case, or\nrequirement.\n\n## Attribution\n\n**MUST READ**: Before executing any `gcloud`, REST API, or MCP command, you\nMUST read and follow the instructions in\n[Usage Attribution](references\u002Fattribution.md).\n\n## Workflow\n\nThe solution design and implementation workflow is divided into the following\nphases:\n\n*   **Phase 1: Requirements discovery and analysis**: Analyze the workload's\n    requirements, constraints, dependencies, and current state.\n*   **Phase 2: Solution design**: Build a technology stack, architecture, and\n    deployment configuration for the workload based on Google Cloud design best\n    practices and recommendations.\n*   **Phase 3: Implementation plan**: Generate automation and instructions to\n    deploy the solution.\n*   **Phase 4: Solution validation**: Validate that the deployment meets the\n    requirements of the workload.\n\nCopy this checklist into your active task\u002Fplan artifact to track progress across\nthe four phases:\n\n-   [ ] Phase 1: Requirements discovery and analysis completed & confirmed.\n-   [ ] Phase 2: Solution architecture generated & approved.\n-   [ ] Phase 3: Implementation plan generated & approved.\n-   [ ] Phase 4: Solution validation generated & approved.\n\n### Phase 1: Requirements discovery and analysis\n\n1.  **Discover requirements**: Gather and understand the functional and\n    non-functional requirements, business goals, and current state (if any) of\n    the workload, including its architecture, dependencies, and constraints.\n\n    *Important*: First, check whether the user's initial prompt has already\n    answered the following questions or whether the prompt explicitly asks you\n    to propose a solution architecture\u002Fdiagram from a given set of parameters.\n\n    -   If the user's prompt provides sufficient requirements and it explicitly\n        requests an architecture proposal or diagram, then skip asking the\n        questions below, and instead proceed to the step **Recommend agent\n        design pattern**.\n    -   If the user's prompt doesn't provide sufficient requirements, then\n        complete these steps to gather missing information:\n\n        1.  Ask the user to describe the functional requirements of their\n            workload: business processes, activities, and use cases.\n\n        2.  Ask the user to describe the non-functional requirements (security,\n            privacy, compliance, reliability, disaster recovery, cost,\n            operations, performance, and sustainability) of their workloads.\n\n        3.  Ask the user what existing systems, knowledge bases, product\n            documentation, or other documentation the AI agents need to access\n            for grounded guidance.\n\n        4.  Ask the user to describe dependencies, if any, on other workloads,\n            products, or tools.\n\n        5.  Review the input that the user has provided so far, and check\n            whether there are any ambiguities or contradictions in the input.\n\n            If you identify any ambiguities or contradictions in the\n            requirements that the user has provided, then do the following for\n            each ambiguity or contradiction that you identify:\n\n            *   Describe the ambiguity or contradiction.\n            *   Ask the user how they wish to resolve the ambiguity or\n                contradiction.\n                *   If the user delegates the choice to you (e.g., the user\n                    replies with \"do what you think is best\" or \"you decide\"),\n                    then provide a clear suggestion to resolve the ambiguity or\n                    contradiction, explain your reasoning, and ask the user to\n                    approve your suggestion.\n\n            **Critical**: Until all the ambiguities and contradictions that you\n            identify are resolved according to the preceding guidance, you must\n            NOT recommend or generate any architecture design, technical\n            decomposition, or Google Cloud product recommendations.\n\n2.  **Recommend agent design pattern**: Evaluate the complexity, workflow,\n    latency, and cost requirements of the workload to recommend an agent design\n    pattern:\n\n    -   **Single-agent system**: Recommend for simpler tasks, acting as an\n        effective starting point to refine core logic and tools.\n    -   **Multi-agent system**: Recommend for complex problems requiring\n        multiple specialized agents to collaborate on a workflow.\n\n3.  **Identify components**: Based on the requirements analysis, generate a\n    technical decomposition of the workload. The technical decomposition must\n    identify the logical components of the workloads and their relationships.\n    Also identify any cross-cloud components, hybrid components, or on-premises\n    components that the solution needs to integrate with.\n\n4.  **Ask for confirmation**: Ask the user to confirm whether the recommended\n    design pattern and technical decomposition match their workload\n    requirements.\n\n5.  **Iterate**: If the user requests changes, generate an updated technical\n    decomposition, and ask the user to confirm the changes. Continue iterating\n    until the user confirms the technical decomposition. Proceed to the next\n    phase only after the user provides confirmation of the technical\n    decomposition.\n\n### Phase 2: Solution design\n\n1.  **Retrieve relevant Google Cloud guidance from\n    `references\u002Frelated-guidance.md`**.\n\n    *Important*: Use the content that you retrieved from\n    `references\u002Frelated-guidance.md` to ground the guidance that you generate in\n    the remaining steps of this phase.\n\n2.  **Map components to Google Cloud products**: For each component in the\n    confirmed technical decomposition, identify the appropriate Google Cloud\n    products and features by consulting\n    [product-mappings.md](references\u002Fproduct-mappings.md) for detailed\n    recommendations, trade-offs, and alternatives across networking, frontends,\n    agent\u002Fmodel runtimes, memory stores, and tools.\n\n3.  **Create architecture diagram**: Create an architecture diagram in Mermaid\n    format: https:\u002F\u002Fgithub.com\u002Fmermaid-js\u002Fmermaid. The diagram should show the\n    components, their relationships, and data\u002Fcontrol flows.\n\n4.  **Generate design recommendations**: Generate design guidance based on the\n    following Google Cloud best practices and recommendations. Use the\n    information in `references\u002Frelated-guidance.md`, with an emphasis on the\n    guidance in `references\u002Fdesign-principles.md`.\n\n5.  **Draft solution architecture**: Compile the requirements, technical\n    decomposition, product mapping, architecture diagram, and design\n    recommendations into a single Markdown file adhering to the format in\n    [solution-template.md](assets\u002Fsolution-template.md). Save this document in\n    the workspace as `solution-architecture.md`.\n\n6.  **Request review**: Present the generated solution architecture (including\n    the complete fenced `mermaid` code block for the diagram) directly to the\n    user in your response, and explicitly request their feedback or approval.\n    When you present the architecture, ask the user to provide approval for you\n    to proceed with an implementation plan.\n\n7.  **Iterate**: If the user requests changes, generate an updated solution\n    architecture and repeat the steps from \"Map components to Google Cloud\n    products\" through \"Request review\" until the user approves the solution\n    architecture.\n\n### Phase 3: Implementation plan\n\n1.  **Retrieve relevant implementation resources**:\n\n    *Important*: Use the resources in references\u002Frelated-guidance.md as the\n    technical foundation for the Infrastructure as Code (IaC) and the deployment\n    instructions that you generate in the remaining steps of this phase.\n\n2.  **Identify deployment prerequisites**: Document prerequisites for the\n    deployment, including the following:\n\n    -   Projects and billing associations\n    -   Required Google Cloud APIs\n    -   Required IAM permissions\n    -   Any other prerequisites\n\n3.  **Generate Infrastructure as Code (IaC)**: Generate code (e.g., Terraform)\n    and deployment scripts to automate the provisioning of the proposed Google\n    Cloud resources.\n\n    -   Where appropriate, alongside or instead of raw infrastructure scripts,\n        instruct the user to set up or enhance the project structure, deployment\n        configuration, and CI\u002FCD pipelines by using Agents CLI commands:\n        -   To create a new project, use `agents-cli create`. Don't use\n            `agents-cli scaffold create`.\n        -   To add deployment and CI\u002FCD scaffolding to an existing repository,\n            use `agents-cli scaffold enhance`.\n\n4.  **Write deployment instructions**: Draft sequential, step-by-step deployment\n    instructions to execute the IaC and initialize the workload components.\n    Compile the deployment prerequisites, IaC, and deployment instructions into\n    a single Markdown file adhering to the format in\n    [implementation-template.md](assets\u002Fimplementation-template.md). Save this\n    document in the workspace as `implementation-instructions.md`.\n\n    -   The instructions MUST provide the exact ADK code to define a stateful\n        agent node that takes a prompt, calls a model, and returns a tool\n        execution request.\n    -   The instructions MUST demonstrate how to register tools like database\n        readers by using Model Context Protocol (MCP) standards.\n    -   If deploying the agent to Cloud Run, the instructions MUST show how to\n        configure Cloud Run to scale to zero when the agent is idle, reducing\n        runtime costs.\n    -   The instructions MUST recommend using encrypted environment variables to\n        store model parameters or private API credentials. Encryption helps to\n        prevent the exposure of sensitive credentials in plain-text container\n        log streams.\n    -   Where appropriate, the instructions MUST specify using the Agents CLI\n        `agents-cli deploy` command (alongside or instead of raw\n        infrastructure\u002Fdeployment scripts) to run the deployment.\n\n5.  **Request review**: Present the generated deployment instructions to the\n    user and explicitly request their feedback and confirmation.\n\n6.  **Iterate**: If the user requests changes, generate an updated\n    implementation plan and repeat the steps from \"Generate Infrastructure as\n    Code (IaC)\" through \"Request review\" until the user approves the\n    implementation plan.\n\n### Phase 4: Solution validation\n\n1.  **Retrieve relevant verification resources**:\n\n    *Important*: Use the resources in references\u002Frelated-guidance.md and their\n    verification patterns as the starting point for the validation checks and\n    verification scripts that you generate in the remaining steps of this phase.\n\n2.  **Define validation checks**: Outline validation steps to verify that the\n    deployed infrastructure meets the workload's requirements:\n\n    -   **Deployment dry-run**: Commands like `terraform plan` to preview\n        changes. Include instructions to run agent deployment in dry-run mode\n        (e.g., using `agents-cli deploy --dry-run` or `-n`) to preview steps and\n        Terraform executions before pushing to production.\n    -   **Local testing and quality verification**: Recommend using the Agents\n        CLI to run and test agent logic locally (`agents-cli run`) and conduct\n        systematic evaluations (`agents-cli eval run`) to verify agent quality\n        and performance before deploying.\n    -   **Connectivity and routing**: Verification of network paths, load\n        balancer routing, and service endpoints.\n    -   **Security policies**: Verification of restricted access, firewall\n        rules, and IAM enforcement.\n\n3.  **Generate verification scripts**: Draft lightweight scripts or command-line\n    instructions (e.g. using `curl`, `gcloud`, or `agents-cli`) that the user\n    can run to perform these validation checks.\n\n    -   The validation plan MUST include instructions using the Agents CLI for\n        local runs, evaluations, and post-deployment validation checks (e.g.,\n        `agents-cli run --url \u003Cservice-url>` to test the deployed service\n        endpoint).\n\n4.  **Compile validation plan**: Document the validation steps, verification\n    scripts, and expected outcomes in a single Markdown file adhering to the\n    format in [validation-template.md](assets\u002Fvalidation-template.md). Save this\n    document in the workspace as `validation-plan.md`.\n\n5.  **Request review**: Present the validation plan to the user and explicitly\n    request their feedback or approval on the validation plan.\n\n6.  **Conduct validation and finalize**: Assist the user in executing the\n    validation checks and troubleshooting any deployment issues. After the\n    solution is validated successfully, request final approval from the user.\n\n7.  **Iterate**: If the user requests changes, generate an updated validation\n    plan and repeat the steps from \"Define validation checks\" through \"Request\n    review\" until the user approves the validation plan.\n\n--------------------------------------------------------------------------------\n\n## References & Supporting Links\n\n*   For the complete list of Google Cloud architectural documentation, product\n    manuals, development kits, and checklists used by this skill, see\n    [related-guidance.md](references\u002Frelated-guidance.md).\n",{"name":5,"metadata":53,"description":7},{"version":54,"category":55},"1.0.2","MultiProductSolutions","main","4c6800d20130b5a3233e414609f1241013cec8cc","skills\u002Fcloud\u002Fgoogle-cloud-solution-build-deploy-agents",2005,"https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fskills","https:\u002F\u002Fgithub.com\u002Fgoogle","https:\u002F\u002Fopensource.google\u002F","Google ❤️ Open Source","2026-11-14 01:51:35",{"source":8,"skillName":5,"commands":66,"prompt":76,"zipAvailable":77},[67,70,73],{"tool":68,"command":69},"npx","npx skills add google\u002Fskills --skill google-cloud-solution-build-deploy-agents",{"tool":71,"command":72},"bunx","bunx skills add google\u002Fskills --skill google-cloud-solution-build-deploy-agents",{"tool":74,"command":75},"pnpm","pnpm dlx skills add google\u002Fskills --skill google-cloud-solution-build-deploy-agents","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fcloud\u002Fgoogle-cloud-solution-build-deploy-agents\u002FSKILL.md",true,[79,85,89,93,99,103,107],{"path":80,"resourceType":81,"resourceTypeDesc":82,"sizeBytes":83,"sourceHtmlUrl":84},"assets\u002Fimplementation-template.md","ASSET","资产文件",800,"https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fcloud\u002Fgoogle-cloud-solution-build-deploy-agents\u002Fassets\u002Fimplementation-template.md",{"path":86,"resourceType":81,"resourceTypeDesc":82,"sizeBytes":87,"sourceHtmlUrl":88},"assets\u002Fsolution-template.md",5027,"https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fcloud\u002Fgoogle-cloud-solution-build-deploy-agents\u002Fassets\u002Fsolution-template.md",{"path":90,"resourceType":81,"resourceTypeDesc":82,"sizeBytes":91,"sourceHtmlUrl":92},"assets\u002Fvalidation-template.md",2162,"https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fcloud\u002Fgoogle-cloud-solution-build-deploy-agents\u002Fassets\u002Fvalidation-template.md",{"path":94,"resourceType":95,"resourceTypeDesc":96,"sizeBytes":97,"sourceHtmlUrl":98},"references\u002Fattribution.md","REFERENCE","参考文档",1971,"https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fcloud\u002Fgoogle-cloud-solution-build-deploy-agents\u002Freferences\u002Fattribution.md",{"path":100,"resourceType":95,"resourceTypeDesc":96,"sizeBytes":101,"sourceHtmlUrl":102},"references\u002Fdesign-principles.md",4011,"https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fcloud\u002Fgoogle-cloud-solution-build-deploy-agents\u002Freferences\u002Fdesign-principles.md",{"path":104,"resourceType":95,"resourceTypeDesc":96,"sizeBytes":105,"sourceHtmlUrl":106},"references\u002Fproduct-mappings.md",10794,"https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fcloud\u002Fgoogle-cloud-solution-build-deploy-agents\u002Freferences\u002Fproduct-mappings.md",{"path":108,"resourceType":95,"resourceTypeDesc":96,"sizeBytes":109,"sourceHtmlUrl":110},"references\u002Frelated-guidance.md",5290,"https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fcloud\u002Fgoogle-cloud-solution-build-deploy-agents\u002Freferences\u002Frelated-guidance.md",[112,146,175,204,231,255],{"skillNo":113,"name":114,"nameZh":6,"description":115,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":116,"categories":117,"keywords":124,"publishedAt":48},"01M6998S2EZKD3GEJ7F943RZTE","sign-in-with-google-web","Implement, configure, and secure Sign In With Google (SiwG) using Google Identity Services (GIS \u002F `https:\u002F\u002Faccounts.google.com\u002Fgsi\u002Fclient`) across web architectures. Use when creating Google sign-in buttons, implementing Google One Tap with FedCM, integrating GIS in React\u002FNext.js\u002FAngular\u002FHTML, verifying ID tokens on backend runtimes (Python\u002FNode.js\u002FGo\u002FJava), enforcing Google Workspace domain restrictions (`hd`), securing client-side ID tokens via WebCrypto nonces or encrypted IndexedDB, embedding in cross-origin iframes via the Intermediate Iframe API (`https:\u002F\u002Faccounts.google.com\u002Fgsi\u002Fintermediate`), configuring Content Security Policy (CSP), COOP, and Permissions Policy (`identity-credentials-get`) headers, or handling sign-out and revocation. Don't use for legacy Google Sign-In (`gapi.auth2`), Google Cloud IAM service accounts, or Android\u002FiOS native Credential Manager.","https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fidentity\u002Fsign-in-with-google-web\u002FSKILL.md",[118,121],{"categoryNo":119,"categoryName":120,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000009","前端开发",{"categoryNo":122,"categoryName":123,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000010","后端与接口",[9,125,126,127,128,129,130,131,132,28,133,134,135,41,136,137,138,139,140,141,142,143,144,145],"sign in with google web","sign","with","web","android","go","security","google workspace","implement","policy","python","react","iam","permissions","buttons","secure","manager","identity","architectures","domain","runtimes",{"skillNo":147,"name":148,"nameZh":6,"description":149,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":150,"categories":151,"keywords":159,"publishedAt":48},"01M6998PHPDCAD684VW5WRG83D","dpop-adoption","Implement and debug OAuth 2.0 DPoP (RFC 9449) refresh token sender-constraining for WebCrypto, Node.js ES6, and browser runtimes integrating with Google's OAuth platform. Use when configuring non-extractable asymmetric key pairs (P-256), generating DPoP Proof JWTs for authorization code exchange and token refresh, or handling 400 use_dpop_nonce challenge retry loops at oauth2.googleapis.com\u002Ftoken. Don't use for unconstrained OAuth 2.0 flows (where refresh tokens are not bound to a client key pair), or for Google Cloud IAM \u002F service account authentication.","https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fidentity\u002Fdpop-adoption\u002FSKILL.md",[152,153,154],{"categoryNo":122,"categoryName":123,"taxonomyType":21,"taxonomyTypeDesc":22},{"categoryNo":24,"categoryName":25,"taxonomyType":21,"taxonomyTypeDesc":22},{"categoryNo":155,"categoryName":156,"taxonomyType":157,"taxonomyTypeDesc":158},"01JBSEEDCAT000000000000030","开发工程师","PROFESSION","适用职业",[160,161,9,28,162,163,164,165,166,167,137,168,169,170,171,172,173,142,145,174,36],"dpop adoption","adoption","es6","retry","authentication","flows","challenge","code","browser","loops","at","node","token","pair","platform",{"skillNo":176,"name":177,"nameZh":6,"description":178,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":179,"categories":180,"keywords":187,"publishedAt":48},"01M6998M72TC73EEAQH4XC7EXW","retrieving-developer-knowledge","Searches, retrieves, and synthesizes official Google developer documentation across Google Cloud, AI\u002FGemini, Android, Chrome, Web, Flutter, Go, Firebase, and other Google developer platforms. Integrates with the Developer Knowledge MCP server (search_documents, get_documents, answer_query) or the Developer Knowledge REST API fallback. Use when searching for gcloud CLI commands, API syntax, IAM permissions, official documentation, architectural comparisons, or product choice overviews. Don't use for local filesystem lookups or non-Google documentation.","https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fdevelopers\u002Fretrieving-developer-knowledge\u002FSKILL.md",[181,182,185,186],{"categoryNo":122,"categoryName":123,"taxonomyType":21,"taxonomyTypeDesc":22},{"categoryNo":183,"categoryName":184,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000021","资料检索与研究",{"categoryNo":24,"categoryName":25,"taxonomyType":21,"taxonomyTypeDesc":22},{"categoryNo":155,"categoryName":156,"taxonomyType":157,"taxonomyTypeDesc":158},[188,189,190,191,28,192,193,194,195,196,197,198,199,36,200,201,202,137,203,138],"retrieving developer knowledge","knowledge","developer","MCP","mcp server","official","developer documentation","get","search","local","choice","documents","query","api","across","documentation",{"skillNo":205,"name":206,"nameZh":6,"description":207,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":208,"categories":209,"keywords":210,"publishedAt":48},"01M6998J2CZEXS9BES9S3WWFKC","migrate-to-rust","Clones, analyzes, plans, and migrates legacy codebases (C++, Fortran, Java, Python, C, Shell) from a GitHub URL or local directory to a modern, high-performance, deterministic, 100% Safe Rust architecture, tracking live telemetry across every task bucket, test failure\u002Ffix cycle, and Profile-Guided Optimization (PGO) stage, and producing a first-person AI storytelling report and interactive HTML infographic. Use when migrating a GitHub repository or local codebase to Rust, porting legacy C++, Fortran, Python, or Java workloads to memory-safe Rust, designing Struct-of-Arrays (SoA) or SIMD Rust architectures, or optimizing Rust ports with LLVM PGO. Don't use for unrelated non-Rust refactoring or single-line syntax questions.","https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fdevelopers\u002Fmigrate-to-rust\u002FSKILL.md",[],[211,212,213,214,215,216,217,218,135,219,220,221,222,223,224,197,225,226,227,36,228,229,202,230],"migrate to rust","rust","migrate","github","memory","performance","directory","safe","ports","llvm","soa","cycle","porting","fix","arrays","infographic","high-performance","URL","architecture","unrelated",{"skillNo":232,"name":233,"nameZh":6,"description":234,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":235,"categories":236,"keywords":240,"publishedAt":48},"01M6998GCC3K29WZ82KXMB9E02","finding-google-skills","Google platform decision and setup guidance, loaded on demand from Google's skill catalog. Use when a developer is choosing or setting up part of their stack, such as where to run a service, a database, storage, messaging, authentication, analytics, ads, or AI model serving, and a Google product is a reasonable candidate - whether or not a vendor is named - or when a request names a Google product or API. Brings in the matching Google skill so the answer can weigh Google options, their trade-offs, and when they are not the right fit. Skip when the stack is already settled on another provider and no Google product is named, or the task involves no platform choice.","https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fdevelopers\u002Ffinding-google-skills\u002FSKILL.md",[237,238,239],{"categoryNo":122,"categoryName":123,"taxonomyType":21,"taxonomyTypeDesc":22},{"categoryNo":24,"categoryName":25,"taxonomyType":21,"taxonomyTypeDesc":22},{"categoryNo":155,"categoryName":156,"taxonomyType":157,"taxonomyTypeDesc":158},[9,241,242,243,244,174,245,246,247,248,249,164,250,251,252,198,37,253,36,254,201],"finding google skills","storage","as","messaging","model","trade-offs","provider","answer","fit","catalog","up","decision","database","candidate",{"skillNo":256,"name":257,"nameZh":6,"description":258,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":259,"categories":260,"keywords":263,"publishedAt":48},"01M6998DWNHTH0BS6W1CY8ZXMP","workload-manager-basics","Use this skill to manage Google Cloud Workload Manager evaluations, rules, scanned resources, and validation results by using public client libraries and the REST API. Use when you need to inspect workload best-practice rules, create and run evaluations for Google Cloud general best practices, SAP, SQL Server, or custom organizational rules, review violations, export results to BigQuery, or automate Workload Manager through client libraries because no service-specific public CLI or MCP server is available. Don't use for general Google Compute Engine instance management, VPC configuration, or standard IAM auditing.","https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fcloud\u002Fworkload-manager-basics\u002FSKILL.md",[261,262],{"categoryNo":122,"categoryName":123,"taxonomyType":21,"taxonomyTypeDesc":22},{"categoryNo":24,"categoryName":25,"taxonomyType":21,"taxonomyTypeDesc":22},[264,265,141,266,9,191,267,28,192,268,269,270,271,36,272,201,137,273,274,275,276],"workload manager basics","workload","basics","cli","review","vpc","resources","inspect","you","create","libraries","server","run"]