[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6991WWEVPXCQREVNT95VP64":3,"skill-same:01M6991WWEVPXCQREVNT95VP64":91},{"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":23,"publishedAt":44,"linkOnly":45,"markdownContent":46,"markdownBody":47,"frontMatter":48,"refName":49,"commitSha":50,"skillDirectory":51,"forksCount":52,"repositoryHtmlUrl":53,"repositoryHomepageUrl":54,"ownerHtmlUrl":55,"ownerBlogUrl":56,"ownerDescription":57,"updatedAt":58,"install":59,"resources":72},"01M6991WWEVPXCQREVNT95VP64","meridian-model-building",null,"Guides users through building a Meridian Marketing Mix Modeling (MMM) model. Use when a user wants to load data, map columns, configure ModelSpec, run Exploratory Data Analysis (EDA), fit a model, and save the model. Don't use for visualizing results or creating a scenario planner.","google\u002Fmeridian","google","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F1342004?v=4","01M698VYEFDM0VJBZDYYJ2T83S","Google","01M6991Q88XBX9SPK8Y92MAEWF",1605,"Apache-2.0","https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fmeridian\u002Fblob\u002Fmain\u002Fskills\u002Fmeridian_model_building\u002FSKILL.md",[18],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000013","分析与可视化","SCENARIO","使用场景",[24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43],"meridian model building","meridian","model","when","data analysis","for","load","mix","scenario planner","users","guides","fit","map","use","results","planner","configure","analysis","exploratory data analysis","marketing mix modeling","2026-11-10 10:12:30",false,"---\nname: meridian-model-building\ndescription: >-\n  Guides users through building a Meridian Marketing Mix Modeling (MMM) model. Use when a user wants to load data, map columns, configure ModelSpec, run Exploratory Data Analysis (EDA), fit a model, and save the model. Don't use for visualizing results or creating a scenario planner.\n---\n\n# Meridian Model Building Skill\n\nThis skill guides the user through the process of creating a Meridian model,\naccumulating the code into a Python script.\n\n## Core Workflow\n\n### Interactivity Checkpoint Rule\n\nThroughout this workflow, you will encounter **CRITICAL INTERACTIVE\nCHECKPOINT**s. At each checkpoint, you MUST:\n\n1.  Present the current proposed configurations, parameters, mappings, script\n    path, or status to the user for approval.\n2.  Ask the user if they are ready to proceed using the available\n    user-interaction tool (e.g., `ask_question`), structured as a\n    multiple-choice question. Do NOT use raw chat text.\n3.  Wait for their response before proceeding.\n    *   *MANDATORY*: You MUST pause at every checkpoint regardless of the\n        initial prompt instructions (even if the user request contains phrases\n        like \"run autonomously\", \"execute directly\", \"fix autonomously\", etc.).\n        The initial request does NOT bypass these interactive checkpoints.\n    *   *Note*: If the user replies to a checkpoint with a generic approval\n        (e.g., \"proceed\", \"do what you think is best\"), proceed with the\n        proposed defaults.\n\n### Python Environment and Meridian Source\n\nBefore running any Python, pick the interpreter of the environment where\n`google-meridian` is installed, and use it for every command and script below:\n\n1.  If the user named an environment, use it. Otherwise, if a virtual\n    environment is already active (`VIRTUAL_ENV` or `CONDA_PREFIX` is set), use\n    `python3`. Otherwise, if the project directory (the current working\n    directory) has a virtual environment (`.venv\u002F` or `venv\u002F`), use its\n    interpreter (e.g., `.venv\u002Fbin\u002Fpython3`). Otherwise use `python3`.\n2.  Check it with `\u003Cinterpreter> -c \"import meridian\"`. If that fails, ask the\n    user which environment to use instead of searching the machine or installing\n    packages unprompted.\n\nSteps below ask you to read Meridian modules (e.g., `meridian.model.spec`) to\nconfirm argument names and defaults. Users usually install Meridian from PyPI\n(`pip install google-meridian`), so its source is not in their project folder.\nFind the installed package and read the files there, or print signatures and\ndocstrings with `inspect`:\n\n```shell\n\u003Cinterpreter> -c \"import meridian, os; print(os.path.dirname(meridian.__file__))\"\n```\n\n--------------------------------------------------------------------------------\n\n### 1. Initial Setup\n\n*   Prompt the user for the input CSV file path, the desired path for the\n    generated Python script, the EDA HTML report output path, and the saved\n    model path (`meridian_model.binpb` by default). **If the user does not\n    specify output paths, default to `model_build\u002F` in the active project\n    directory (or relative to the input data directory) for the script and all\n    outputs (`meridian_model.binpb`, `eda.html`).**\n*   **CRITICAL INTERACTIVE CHECKPOINT**: Present the gathered paths to the user\n    and obtain confirmation before proceeding to data loading.\n\n### 2. Add Data Loading & Column Mapping Code\n\n*   **Target Module:** `meridian.data.data_frame_input_data_builder`\n*   **Action:**\n    *   **Check CSV Format**: Before loading data, verify if the CSV data is in\n        the right format. Use the `meridian-doc-consultant` skill to find the\n        data-loading guide on `developers.google.com\u002Fmeridian` that matches the\n        **columns observed in the data** (e.g., the \"load geo-level data without\n        reach and frequency\" or \"load geo-level data with organic and non-media\"\n        guide) to understand the expected columns and data types. Consult\n        `references\u002Fcsv_format_reference.md` for details on expected row\u002Fcolumn\n        structure and data quality guardrails. If the format is incorrect or\n        missing required columns, **attempt to autonomously convert the dataset\n        to the expected format for the user** (e.g., renaming columns,\n        restructuring) unless you are uncertain and need user input.\n    *   Read the header row of the provided CSV using Python to get the column\n        names.\n    *   Propose heuristic mappings based on column keywords (e.g., 'sales' ->\n        `kpi_col`, 'spend' -> `media_spend_cols`) and infer the `kpi_type`\n        ('revenue' or 'non_revenue') based on the columns (e.g., 'revenue' or\n        'sales' implying 'revenue', and 'conversions' or 'leads' implying\n        'non_revenue').\n    *   **Robust Mapping**: If the user prompt specifies mapping a column name\n        that does not exist in the CSV, do not assume it is a literal name if it\n        looks like a description (e.g., 'media_impressions' vs\n        'ChannelX_impression'). Use heuristics to find matching columns and\n        proceed.\n    *   Present the proposed mapping to the user.\n    *   **CRITICAL INTERACTIVE CHECKPOINT**: Present the proposed column\n        mappings to the user and obtain approval before continuing to model\n        configuration.\n    *   Accumulate the data loading code using\n        `meridian.data.data_frame_input_data_builder.DataFrameInputDataBuilder`\n        and its `with_*` methods (e.g. `with_kpi`, `with_media`). See\n        [data_builder_template.md](references\u002Fdata_builder_template.md).\n\n### 3. Add Model Configuration Code\n\n*   **Target Modules:** `meridian.model.spec`, `meridian.model.model`\n*   **Action:**\n    *   Read the `ModelSpec` and `PriorDistribution` definitions in\n        `meridian.model.spec`.\n    *   Guide the user through configuration, prompting for relevant values\n        while explaining their purpose based on the source code docstrings.\n    *   **CRITICAL INTERACTIVE CHECKPOINT**: Present the proposed model\n        specification parameters to the user and obtain approval before\n        continuing.\n    *   Accumulate the code to initialize `meridian.model.spec.ModelSpec` and\n        `meridian.model.model.Meridian`. See\n        [model_spec_template.md](references\u002Fmodel_spec_template.md).\n    *   Accumulate code: `mmm.sample_prior()`\n\n### 4. Add Exploratory Data Analysis (EDA) Code\n\n*   **Target Module:** `meridian.model.eda.meridian_eda`\n*   **Action:**\n    *   Read `meridian_eda.py` or module docstrings to confirm the\n        `generate_and_save_report` method.\n    *   Accumulate code to initialize `meridian_eda.MeridianEDA` and call\n        `generate_and_save_report(filepath)` using the user's specified path.\n    *   **CRITICAL INTERACTIVE CHECKPOINT**: Present the EDA output path\n        configuration and obtain approval before proceeding to the model fitting\n        step.\n\n### 5. Add Model Fitting Code\n\n*   **Target Module:** `meridian.model.model`\n*   **Action:**\n    *   Read the `sample_posterior` method in `meridian.model.model` to\n        understand its parameters.\n    *   Prompt the user for MCMC parameters: `n_chains`, `n_adapt`, `n_burnin`,\n        `n_keep`.\n    *   **CRITICAL INTERACTIVE CHECKPOINT**: Present the MCMC parameters to the\n        user and obtain approval before proceeding to compile the model fitting\n        code.\n    *   Accumulate code: `mmm.sample_posterior(...)`\n\n### 6. Add Model Saving Code\n\n*   **Target Module:** `meridian.schema.serde.meridian_serde`\n*   **Action:**\n    *   Generate code to save the model using `meridian_serde.save_meridian()`\n        to the user-specified path (or the default). See\n        [script_template.md](references\u002Fscript_template.md).\n    *   **Default Filename**: The default filename for the saved model is\n        `meridian_model.binpb` (in the `model_build\u002F` directory). Use this\n        filename if the user does not specify a model filename, even if the\n        script file is named differently.\n    *   **Skip Sampling Handling**: If the user requests to skip fitting or\n        posterior sampling, still include the model saving step\n        (`meridian_serde.save_meridian(mmm, save_path)`) using the initialized\n        `Meridian` model object so the output model file is always created.\n    *   **WARNING:** Do NOT use the deprecated `meridian.model.model.save_mmm`\n        function. Use `meridian_serde.save_meridian` exclusively.\n    *   **CRITICAL INTERACTIVE CHECKPOINT**: Present the model save path and\n        filename to the user and obtain approval before proceeding to script\n        execution.\n\n### 7. Execution & Script Setup\n\n*   **Action:**\n    *   Write the accumulated Python script to the user-specified path. If your\n        file-writing tool can ask the user to review the written file, don't\n        request that review: the next checkpoint already asks for approval, and\n        an extra prompt pauses the workflow.\n    *   **CRITICAL INTERACTIVE CHECKPOINT**: Ask the user for final confirmation\n        to execute the model building script now.\n    *   **Artifact Preservation**: When completing a task that requires\n        generating outputs (like scripts, models, or reports), do **NOT** delete\n        these generated artifacts at the end of your turn. They are the\n        deliverables requested by the user. Only clean up truly temporary\n        scratch files if necessary.\n        *   **Path Handling for Outputs**: In generated scripts, use the output\n            paths confirmed in Step 1, relative to the project directory the\n            script is run from (or absolute paths).\n    *   **Execute the Script**:\n        *   Always run the script with the project directory as the working\n            directory (not a subdirectory), so relative paths resolve as\n            confirmed.\n        *   Use the interpreter chosen in \"Python Environment and Meridian\n            Source\" above.\n        *   Example command: `.venv\u002Fbin\u002Fpython3 model_build\u002Fmy_model.py`\n    *   **Handling Long Runs**: If the command is sent to the background due to\n        execution time, wait for the background task to complete and check the\n        final output to catch runtime errors.\n    *   **Differentiated Error Handling**:\n        *   If it's a **Syntax Error** or **Import Error**, read the relevant\n            source code to understand the correct usage or interface.\n        *   If it's a **ValueError** or parameter constraint violation (e.g.,\n            `knots` too large), check the **docstring** of the class\u002Ffunction or\n            consult the `meridian-doc-consultant` skill to find valid values in\n            the documentation.\n        *   *Autonomy*: If a fix requires changing configuration, prompt the\n            user for confirmation. If the user response grants autonomy, proceed\n            to fix it.\n\n### 8. Conclusion\n\n*   **Action:**\n    *   Confirm execution success and artifact generation.\n","# Meridian Model Building Skill\n\nThis skill guides the user through the process of creating a Meridian model,\naccumulating the code into a Python script.\n\n## Core Workflow\n\n### Interactivity Checkpoint Rule\n\nThroughout this workflow, you will encounter **CRITICAL INTERACTIVE\nCHECKPOINT**s. At each checkpoint, you MUST:\n\n1.  Present the current proposed configurations, parameters, mappings, script\n    path, or status to the user for approval.\n2.  Ask the user if they are ready to proceed using the available\n    user-interaction tool (e.g., `ask_question`), structured as a\n    multiple-choice question. Do NOT use raw chat text.\n3.  Wait for their response before proceeding.\n    *   *MANDATORY*: You MUST pause at every checkpoint regardless of the\n        initial prompt instructions (even if the user request contains phrases\n        like \"run autonomously\", \"execute directly\", \"fix autonomously\", etc.).\n        The initial request does NOT bypass these interactive checkpoints.\n    *   *Note*: If the user replies to a checkpoint with a generic approval\n        (e.g., \"proceed\", \"do what you think is best\"), proceed with the\n        proposed defaults.\n\n### Python Environment and Meridian Source\n\nBefore running any Python, pick the interpreter of the environment where\n`google-meridian` is installed, and use it for every command and script below:\n\n1.  If the user named an environment, use it. Otherwise, if a virtual\n    environment is already active (`VIRTUAL_ENV` or `CONDA_PREFIX` is set), use\n    `python3`. Otherwise, if the project directory (the current working\n    directory) has a virtual environment (`.venv\u002F` or `venv\u002F`), use its\n    interpreter (e.g., `.venv\u002Fbin\u002Fpython3`). Otherwise use `python3`.\n2.  Check it with `\u003Cinterpreter> -c \"import meridian\"`. If that fails, ask the\n    user which environment to use instead of searching the machine or installing\n    packages unprompted.\n\nSteps below ask you to read Meridian modules (e.g., `meridian.model.spec`) to\nconfirm argument names and defaults. Users usually install Meridian from PyPI\n(`pip install google-meridian`), so its source is not in their project folder.\nFind the installed package and read the files there, or print signatures and\ndocstrings with `inspect`:\n\n```shell\n\u003Cinterpreter> -c \"import meridian, os; print(os.path.dirname(meridian.__file__))\"\n```\n\n--------------------------------------------------------------------------------\n\n### 1. Initial Setup\n\n*   Prompt the user for the input CSV file path, the desired path for the\n    generated Python script, the EDA HTML report output path, and the saved\n    model path (`meridian_model.binpb` by default). **If the user does not\n    specify output paths, default to `model_build\u002F` in the active project\n    directory (or relative to the input data directory) for the script and all\n    outputs (`meridian_model.binpb`, `eda.html`).**\n*   **CRITICAL INTERACTIVE CHECKPOINT**: Present the gathered paths to the user\n    and obtain confirmation before proceeding to data loading.\n\n### 2. Add Data Loading & Column Mapping Code\n\n*   **Target Module:** `meridian.data.data_frame_input_data_builder`\n*   **Action:**\n    *   **Check CSV Format**: Before loading data, verify if the CSV data is in\n        the right format. Use the `meridian-doc-consultant` skill to find the\n        data-loading guide on `developers.google.com\u002Fmeridian` that matches the\n        **columns observed in the data** (e.g., the \"load geo-level data without\n        reach and frequency\" or \"load geo-level data with organic and non-media\"\n        guide) to understand the expected columns and data types. Consult\n        `references\u002Fcsv_format_reference.md` for details on expected row\u002Fcolumn\n        structure and data quality guardrails. If the format is incorrect or\n        missing required columns, **attempt to autonomously convert the dataset\n        to the expected format for the user** (e.g., renaming columns,\n        restructuring) unless you are uncertain and need user input.\n    *   Read the header row of the provided CSV using Python to get the column\n        names.\n    *   Propose heuristic mappings based on column keywords (e.g., 'sales' ->\n        `kpi_col`, 'spend' -> `media_spend_cols`) and infer the `kpi_type`\n        ('revenue' or 'non_revenue') based on the columns (e.g., 'revenue' or\n        'sales' implying 'revenue', and 'conversions' or 'leads' implying\n        'non_revenue').\n    *   **Robust Mapping**: If the user prompt specifies mapping a column name\n        that does not exist in the CSV, do not assume it is a literal name if it\n        looks like a description (e.g., 'media_impressions' vs\n        'ChannelX_impression'). Use heuristics to find matching columns and\n        proceed.\n    *   Present the proposed mapping to the user.\n    *   **CRITICAL INTERACTIVE CHECKPOINT**: Present the proposed column\n        mappings to the user and obtain approval before continuing to model\n        configuration.\n    *   Accumulate the data loading code using\n        `meridian.data.data_frame_input_data_builder.DataFrameInputDataBuilder`\n        and its `with_*` methods (e.g. `with_kpi`, `with_media`). See\n        [data_builder_template.md](references\u002Fdata_builder_template.md).\n\n### 3. Add Model Configuration Code\n\n*   **Target Modules:** `meridian.model.spec`, `meridian.model.model`\n*   **Action:**\n    *   Read the `ModelSpec` and `PriorDistribution` definitions in\n        `meridian.model.spec`.\n    *   Guide the user through configuration, prompting for relevant values\n        while explaining their purpose based on the source code docstrings.\n    *   **CRITICAL INTERACTIVE CHECKPOINT**: Present the proposed model\n        specification parameters to the user and obtain approval before\n        continuing.\n    *   Accumulate the code to initialize `meridian.model.spec.ModelSpec` and\n        `meridian.model.model.Meridian`. See\n        [model_spec_template.md](references\u002Fmodel_spec_template.md).\n    *   Accumulate code: `mmm.sample_prior()`\n\n### 4. Add Exploratory Data Analysis (EDA) Code\n\n*   **Target Module:** `meridian.model.eda.meridian_eda`\n*   **Action:**\n    *   Read `meridian_eda.py` or module docstrings to confirm the\n        `generate_and_save_report` method.\n    *   Accumulate code to initialize `meridian_eda.MeridianEDA` and call\n        `generate_and_save_report(filepath)` using the user's specified path.\n    *   **CRITICAL INTERACTIVE CHECKPOINT**: Present the EDA output path\n        configuration and obtain approval before proceeding to the model fitting\n        step.\n\n### 5. Add Model Fitting Code\n\n*   **Target Module:** `meridian.model.model`\n*   **Action:**\n    *   Read the `sample_posterior` method in `meridian.model.model` to\n        understand its parameters.\n    *   Prompt the user for MCMC parameters: `n_chains`, `n_adapt`, `n_burnin`,\n        `n_keep`.\n    *   **CRITICAL INTERACTIVE CHECKPOINT**: Present the MCMC parameters to the\n        user and obtain approval before proceeding to compile the model fitting\n        code.\n    *   Accumulate code: `mmm.sample_posterior(...)`\n\n### 6. Add Model Saving Code\n\n*   **Target Module:** `meridian.schema.serde.meridian_serde`\n*   **Action:**\n    *   Generate code to save the model using `meridian_serde.save_meridian()`\n        to the user-specified path (or the default). See\n        [script_template.md](references\u002Fscript_template.md).\n    *   **Default Filename**: The default filename for the saved model is\n        `meridian_model.binpb` (in the `model_build\u002F` directory). Use this\n        filename if the user does not specify a model filename, even if the\n        script file is named differently.\n    *   **Skip Sampling Handling**: If the user requests to skip fitting or\n        posterior sampling, still include the model saving step\n        (`meridian_serde.save_meridian(mmm, save_path)`) using the initialized\n        `Meridian` model object so the output model file is always created.\n    *   **WARNING:** Do NOT use the deprecated `meridian.model.model.save_mmm`\n        function. Use `meridian_serde.save_meridian` exclusively.\n    *   **CRITICAL INTERACTIVE CHECKPOINT**: Present the model save path and\n        filename to the user and obtain approval before proceeding to script\n        execution.\n\n### 7. Execution & Script Setup\n\n*   **Action:**\n    *   Write the accumulated Python script to the user-specified path. If your\n        file-writing tool can ask the user to review the written file, don't\n        request that review: the next checkpoint already asks for approval, and\n        an extra prompt pauses the workflow.\n    *   **CRITICAL INTERACTIVE CHECKPOINT**: Ask the user for final confirmation\n        to execute the model building script now.\n    *   **Artifact Preservation**: When completing a task that requires\n        generating outputs (like scripts, models, or reports), do **NOT** delete\n        these generated artifacts at the end of your turn. They are the\n        deliverables requested by the user. Only clean up truly temporary\n        scratch files if necessary.\n        *   **Path Handling for Outputs**: In generated scripts, use the output\n            paths confirmed in Step 1, relative to the project directory the\n            script is run from (or absolute paths).\n    *   **Execute the Script**:\n        *   Always run the script with the project directory as the working\n            directory (not a subdirectory), so relative paths resolve as\n            confirmed.\n        *   Use the interpreter chosen in \"Python Environment and Meridian\n            Source\" above.\n        *   Example command: `.venv\u002Fbin\u002Fpython3 model_build\u002Fmy_model.py`\n    *   **Handling Long Runs**: If the command is sent to the background due to\n        execution time, wait for the background task to complete and check the\n        final output to catch runtime errors.\n    *   **Differentiated Error Handling**:\n        *   If it's a **Syntax Error** or **Import Error**, read the relevant\n            source code to understand the correct usage or interface.\n        *   If it's a **ValueError** or parameter constraint violation (e.g.,\n            `knots` too large), check the **docstring** of the class\u002Ffunction or\n            consult the `meridian-doc-consultant` skill to find valid values in\n            the documentation.\n        *   *Autonomy*: If a fix requires changing configuration, prompt the\n            user for confirmation. If the user response grants autonomy, proceed\n            to fix it.\n\n### 8. Conclusion\n\n*   **Action:**\n    *   Confirm execution success and artifact generation.\n",{"name":5,"description":7},"main","271d6c9316f160090adcfe2e4b73dcce78a3edb0","skills\u002Fmeridian_model_building",312,"https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fmeridian","https:\u002F\u002Fdevelopers.google.com\u002Fmeridian","https:\u002F\u002Fgithub.com\u002Fgoogle","https:\u002F\u002Fopensource.google\u002F","Google ❤️ Open Source","2026-11-14 11:56:01",{"source":8,"skillName":5,"commands":60,"prompt":70,"zipAvailable":71},[61,64,67],{"tool":62,"command":63},"npx","npx skills add google\u002Fmeridian --skill meridian-model-building",{"tool":65,"command":66},"bunx","bunx skills add google\u002Fmeridian --skill meridian-model-building",{"tool":68,"command":69},"pnpm","pnpm dlx skills add google\u002Fmeridian --skill meridian-model-building","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fmeridian\u002Fblob\u002Fmain\u002Fskills\u002Fmeridian_model_building\u002FSKILL.md",true,[73,79,83,87],{"path":74,"resourceType":75,"resourceTypeDesc":76,"sizeBytes":77,"sourceHtmlUrl":78},"references\u002Fcsv_format_reference.md","REFERENCE","参考文档",2351,"https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fmeridian\u002Fblob\u002Fmain\u002Fskills\u002Fmeridian_model_building\u002Freferences\u002Fcsv_format_reference.md",{"path":80,"resourceType":75,"resourceTypeDesc":76,"sizeBytes":81,"sourceHtmlUrl":82},"references\u002Fdata_builder_template.md",3254,"https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fmeridian\u002Fblob\u002Fmain\u002Fskills\u002Fmeridian_model_building\u002Freferences\u002Fdata_builder_template.md",{"path":84,"resourceType":75,"resourceTypeDesc":76,"sizeBytes":85,"sourceHtmlUrl":86},"references\u002Fmodel_spec_template.md",2620,"https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fmeridian\u002Fblob\u002Fmain\u002Fskills\u002Fmeridian_model_building\u002Freferences\u002Fmodel_spec_template.md",{"path":88,"resourceType":75,"resourceTypeDesc":76,"sizeBytes":89,"sourceHtmlUrl":90},"references\u002Fscript_template.md",2141,"https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fmeridian\u002Fblob\u002Fmain\u002Fskills\u002Fmeridian_model_building\u002Freferences\u002Fscript_template.md",[92,118,133,152],{"skillNo":93,"name":94,"nameZh":6,"description":95,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":96,"categories":97,"keywords":104,"publishedAt":44},"01M6992184V3ZDYFMX8R0Z2226","meridian-scenario-planner","Generates Scenario Planner data from a fitted Meridian model and exports it for Looker Studio dashboard creation via Colab handoff. Use when the user wants to generate scenario planning data, budget grids, or Looker Studio dashboards from a saved model. Don't use for visualizing model results (use meridian-result-visualization) or running standard budget optimization (use meridian-budget-optimization).","https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fmeridian\u002Fblob\u002Fmain\u002Fskills\u002Fmeridian_scenario_planner\u002FSKILL.md",[98,99],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},{"categoryNo":100,"categoryName":101,"taxonomyType":102,"taxonomyTypeDesc":103},"01JBSEEDCAT000000000000033","数据分析师","PROFESSION","适用职业",[105,25,39,106,32,27,107,108,109,110,29,111,112,113,114,115,26,116,38,37,117],"meridian scenario planner","scenario","running","standard","via","handoff","dashboard","saved","visualization","planning","Optimization","studio","creation",{"skillNo":119,"name":120,"nameZh":6,"description":121,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":122,"categories":123,"keywords":125,"publishedAt":44},"01M6991ZTBAGT71YK8FYA4RKVT","meridian-result-visualization","Loads a fitted Meridian model and generates standard visualization reports: Model Results Summary. Use when the user wants to load a saved model and view results (model performance, fit, health). Don't use for running budget optimization (use meridian-budget-optimization) or building\u002Ffitting the model (use meridian-model-building).","https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fmeridian\u002Fblob\u002Fmain\u002Fskills\u002Fmeridian_result_visualization\u002FSKILL.md",[124],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},[113,25,126,127,38,37,128,27,35,129,112,30,130,29,26,108,107,115,131,132],"meridian result visualization","result","reports","loads","view","performance","health",{"skillNo":134,"name":135,"nameZh":6,"description":136,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":137,"categories":138,"keywords":139,"publishedAt":44},"01M6991VAJ8WFE8KBKNTM6H16F","meridian-doc-consultant","Assists users with questions about Meridian Marketing Mix Modeling and Meridian GeoX geo-experiment concepts, parameters, and best practices by finding and consulting the relevant documentation on developers.google.com\u002Fmeridian. Use when the user asks for explanations or guidance on Meridian MMM or GeoX topics. Don't use for writing code or running models directly.","https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fmeridian\u002Fblob\u002Fmain\u002Fskills\u002Fmeridian_doc_consultant\u002FSKILL.md",[],[140,141,25,142,9,143,43,144,107,27,145,29,31,146,33,147,148,37,149,150,151],"consultant","doc","meridian doc consultant","writing","relevant","topics","with","geo","models","best practices","code","documentation",{"skillNo":153,"name":154,"nameZh":6,"description":155,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":156,"categories":157,"keywords":161,"publishedAt":44},"01M6991SZPBFVMX04EDK0YH5RV","meridian-budget-optimization","Loads a fitted Meridian model and runs budget optimization scenarios. Use when the user wants to load a saved model and find the optimal budget allocation or target ROI spending. Don't use for visualizing model performance (use meridian-result-visualization) or building\u002Ffitting the model (use meridian-model-building).","https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fmeridian\u002Fblob\u002Fmain\u002Fskills\u002Fmeridian_budget_optimization\u002FSKILL.md",[158],{"categoryNo":159,"categoryName":160,"taxonomyType":102,"taxonomyTypeDesc":103},"01JBSEEDCAT000000000000040","财务",[162,163,115,164,25,165,27,127,29,166,30,112,167,129,37,168,26,113,169,131],"meridian budget optimization","budget optimization","Budget","spending","allocation","budget allocation","find","roi"]