[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6991VAJ8WFE8KBKNTM6H16F":3,"skill-same:01M6991VAJ8WFE8KBKNTM6H16F":74},{"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":18,"publishedAt":39,"linkOnly":40,"markdownContent":41,"markdownBody":42,"frontMatter":43,"refName":44,"commitSha":45,"skillDirectory":46,"forksCount":47,"repositoryHtmlUrl":48,"repositoryHomepageUrl":49,"ownerHtmlUrl":50,"ownerBlogUrl":51,"ownerDescription":52,"updatedAt":53,"install":54,"resources":67},"01M6991VAJ8WFE8KBKNTM6H16F","meridian-doc-consultant",null,"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.","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_doc_consultant\u002FSKILL.md",[],[19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,9],"meridian doc consultant","meridian","doc","consultant","models","running","when","topics","for","mix","with","relevant","users","geo","writing","use","best practices","code","documentation","marketing mix modeling","2026-11-10 10:12:30",false,"---\nname: meridian-doc-consultant\ndescription: >-\n  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.\n---\n\n# Meridian Documentation Consultant\n\nThis skill guides the agent in finding and consulting the right page of the\nofficial Meridian documentation to answer user questions token-efficiently.\n\nThe documentation is published at `https:\u002F\u002Fdevelopers.google.com\u002Fmeridian\u002F`. It\ncovers both Meridian MMM (`\u002Fmeridian\u002Fdocs\u002F...`) and Meridian GeoX geo\nexperiments (`\u002Fmeridian\u002Fgeox\u002F...`). It is **not** included in the Meridian\nGitHub repository or the `google-meridian` Python package, so do not look for\ndocumentation files on disk.\n\n## Core Workflow\n\nWhen the user asks a conceptual or procedural question about Meridian, follow\nthese steps:\n\n### 1. Identify the Topic\n\n*   Analyze the user's question to identify the core topic (e.g., \"knots\",\n    \"priors\", \"data format\").\n\n### 2. Locate Candidate Pages (RAG-style Retrieval)\n\n*   Consult the [Documentation Map](references\u002Fdocumentation_map.md) and use its\n    summaries to identify a **batch of potentially relevant pages** (e.g., 3-5\n    pages) related to the user's query.\n*   **Direct Term Search**: If the user asks about specific technical terms\n    (e.g., `eta_m`, `xi_c`, `knots`), search the Documentation Map for these\n    terms and then search the text of the candidate pages you fetch in step 3,\n    **EVEN IF** a page in the Documentation Map seems to cover the topic. If the\n    map has no match, you may use web search restricted to\n    `developers.google.com\u002Fmeridian` to find candidate pages.\n*   **Concept Intersection**: If the query involves multiple concepts (e.g.,\n    \"priors\" + \"insufficient data\"), look for pages that discuss these concepts\n    together.\n*   **CRITICAL**: Do NOT fetch pages one by one to discover the right one, as\n    fetching unrelated pages wastes tokens and latency. Rely on the map's\n    summaries to identify candidates.\n\n### 3. Fetch and Read Candidates\n\n*   From the candidate list, select the **Top N (e.g., 3) most promising pages**\n    to read.\n*   Fetch the **Markdown version** of each page by appending `.md.txt` to its\n    URL. For example, the page\n    `https:\u002F\u002Fdevelopers.google.com\u002Fmeridian\u002Fdocs\u002Fpre-modeling\u002Famount-data-needed`\n    is available as plain Markdown at\n    `https:\u002F\u002Fdevelopers.google.com\u002Fmeridian\u002Fdocs\u002Fpre-modeling\u002Famount-data-needed.md.txt`.\n    Use your URL-fetching tool, or `curl -sL \u003Curl>.md.txt` if you only have a\n    shell. The regular HTML page is mostly site layout, so prefer the `.md.txt`\n    version.\n*   For large pages, search the fetched text for specific keywords to locate the\n    relevant section instead of reading everything.\n*   **Broaden Search Terms**: If a specific keyword search fails (e.g., \"maximum\n    channels\"), try searching for broader related concepts (e.g., \"channels\") to\n    find the relevant section.\n*   **Searching Recommendations**: When looking for recommendations, search for\n    keywords like \"recommend\", \"best practice\", or \"should\" combined with the\n    topic keyword.\n*   **Handling Fuzzy Matches on Limits**: If the user asks for a \"maximum\" and\n    the docs say \"below X\", provide that value.\n*   **Prefer Documentation over Code**: Prioritize the documentation pages over\n    source code files for conceptual questions.\n*   **No Web Access**: If you cannot fetch the pages, tell the user that this\n    skill needs access to `developers.google.com\u002Fmeridian`, give them the page\n    URLs from the Documentation Map to read, and do not present unverified\n    details as documented guidance.\n\n### 4. Synthesize and Answer\n\n*   **Multi-Document Synthesis**: The answer does not have to depend on a single\n    page. Synthesize the answer from multiple candidate pages if they provide\n    complementary information.\n*   **Grounding Constraint**: Do not answer with specific numbers or\n    recommendations from your general knowledge or from search-result snippets\n    if they are not present in a page you fetched and read.\n*   **Cite the Source**: In your final response, you **MUST** cite the full page\n    URL(s) (e.g.,\n    `https:\u002F\u002Fdevelopers.google.com\u002Fmeridian\u002Fdocs\u002Fpre-modeling\u002Famount-data-needed`)\n    of all documentation you consulted. Only cite pages you actually fetched and\n    read. Do not cite local file paths such as `docs\u002F...md`; they do not exist\n    for Meridian users. Ensure you cite the specific page that provided the\n    answer to the specific condition.\n","# Meridian Documentation Consultant\n\nThis skill guides the agent in finding and consulting the right page of the\nofficial Meridian documentation to answer user questions token-efficiently.\n\nThe documentation is published at `https:\u002F\u002Fdevelopers.google.com\u002Fmeridian\u002F`. It\ncovers both Meridian MMM (`\u002Fmeridian\u002Fdocs\u002F...`) and Meridian GeoX geo\nexperiments (`\u002Fmeridian\u002Fgeox\u002F...`). It is **not** included in the Meridian\nGitHub repository or the `google-meridian` Python package, so do not look for\ndocumentation files on disk.\n\n## Core Workflow\n\nWhen the user asks a conceptual or procedural question about Meridian, follow\nthese steps:\n\n### 1. Identify the Topic\n\n*   Analyze the user's question to identify the core topic (e.g., \"knots\",\n    \"priors\", \"data format\").\n\n### 2. Locate Candidate Pages (RAG-style Retrieval)\n\n*   Consult the [Documentation Map](references\u002Fdocumentation_map.md) and use its\n    summaries to identify a **batch of potentially relevant pages** (e.g., 3-5\n    pages) related to the user's query.\n*   **Direct Term Search**: If the user asks about specific technical terms\n    (e.g., `eta_m`, `xi_c`, `knots`), search the Documentation Map for these\n    terms and then search the text of the candidate pages you fetch in step 3,\n    **EVEN IF** a page in the Documentation Map seems to cover the topic. If the\n    map has no match, you may use web search restricted to\n    `developers.google.com\u002Fmeridian` to find candidate pages.\n*   **Concept Intersection**: If the query involves multiple concepts (e.g.,\n    \"priors\" + \"insufficient data\"), look for pages that discuss these concepts\n    together.\n*   **CRITICAL**: Do NOT fetch pages one by one to discover the right one, as\n    fetching unrelated pages wastes tokens and latency. Rely on the map's\n    summaries to identify candidates.\n\n### 3. Fetch and Read Candidates\n\n*   From the candidate list, select the **Top N (e.g., 3) most promising pages**\n    to read.\n*   Fetch the **Markdown version** of each page by appending `.md.txt` to its\n    URL. For example, the page\n    `https:\u002F\u002Fdevelopers.google.com\u002Fmeridian\u002Fdocs\u002Fpre-modeling\u002Famount-data-needed`\n    is available as plain Markdown at\n    `https:\u002F\u002Fdevelopers.google.com\u002Fmeridian\u002Fdocs\u002Fpre-modeling\u002Famount-data-needed.md.txt`.\n    Use your URL-fetching tool, or `curl -sL \u003Curl>.md.txt` if you only have a\n    shell. The regular HTML page is mostly site layout, so prefer the `.md.txt`\n    version.\n*   For large pages, search the fetched text for specific keywords to locate the\n    relevant section instead of reading everything.\n*   **Broaden Search Terms**: If a specific keyword search fails (e.g., \"maximum\n    channels\"), try searching for broader related concepts (e.g., \"channels\") to\n    find the relevant section.\n*   **Searching Recommendations**: When looking for recommendations, search for\n    keywords like \"recommend\", \"best practice\", or \"should\" combined with the\n    topic keyword.\n*   **Handling Fuzzy Matches on Limits**: If the user asks for a \"maximum\" and\n    the docs say \"below X\", provide that value.\n*   **Prefer Documentation over Code**: Prioritize the documentation pages over\n    source code files for conceptual questions.\n*   **No Web Access**: If you cannot fetch the pages, tell the user that this\n    skill needs access to `developers.google.com\u002Fmeridian`, give them the page\n    URLs from the Documentation Map to read, and do not present unverified\n    details as documented guidance.\n\n### 4. Synthesize and Answer\n\n*   **Multi-Document Synthesis**: The answer does not have to depend on a single\n    page. Synthesize the answer from multiple candidate pages if they provide\n    complementary information.\n*   **Grounding Constraint**: Do not answer with specific numbers or\n    recommendations from your general knowledge or from search-result snippets\n    if they are not present in a page you fetched and read.\n*   **Cite the Source**: In your final response, you **MUST** cite the full page\n    URL(s) (e.g.,\n    `https:\u002F\u002Fdevelopers.google.com\u002Fmeridian\u002Fdocs\u002Fpre-modeling\u002Famount-data-needed`)\n    of all documentation you consulted. Only cite pages you actually fetched and\n    read. Do not cite local file paths such as `docs\u002F...md`; they do not exist\n    for Meridian users. Ensure you cite the specific page that provided the\n    answer to the specific condition.\n",{"name":5,"description":7},"main","271d6c9316f160090adcfe2e4b73dcce78a3edb0","skills\u002Fmeridian_doc_consultant",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":55,"prompt":65,"zipAvailable":66},[56,59,62],{"tool":57,"command":58},"npx","npx skills add google\u002Fmeridian --skill meridian-doc-consultant",{"tool":60,"command":61},"bunx","bunx skills add google\u002Fmeridian --skill meridian-doc-consultant",{"tool":63,"command":64},"pnpm","pnpm dlx skills add google\u002Fmeridian --skill meridian-doc-consultant","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fmeridian\u002Fblob\u002Fmain\u002Fskills\u002Fmeridian_doc_consultant\u002FSKILL.md",true,[68],{"path":69,"resourceType":70,"resourceTypeDesc":71,"sizeBytes":72,"sourceHtmlUrl":73},"references\u002Fdocumentation_map.md","REFERENCE","参考文档",71495,"https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fmeridian\u002Fblob\u002Fmain\u002Fskills\u002Fmeridian_doc_consultant\u002Freferences\u002Fdocumentation_map.md",[75,108,125,140],{"skillNo":76,"name":77,"nameZh":6,"description":78,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":79,"categories":80,"keywords":91,"publishedAt":39},"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",[81,86],{"categoryNo":82,"categoryName":83,"taxonomyType":84,"taxonomyTypeDesc":85},"01JBSEEDCAT000000000000013","分析与可视化","SCENARIO","使用场景",{"categoryNo":87,"categoryName":88,"taxonomyType":89,"taxonomyTypeDesc":90},"01JBSEEDCAT000000000000033","数据分析师","PROFESSION","适用职业",[92,20,93,94,95,96,25,24,97,98,99,27,100,101,102,103,104,105,34,106,107],"meridian scenario planner","planner","scenario","scenario planner","standard","via","handoff","dashboard","saved","visualization","planning","Optimization","model","results","creation","studio",{"skillNo":109,"name":110,"nameZh":6,"description":111,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":112,"categories":113,"keywords":115,"publishedAt":39},"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",[114],{"categoryNo":82,"categoryName":83,"taxonomyType":84,"taxonomyTypeDesc":85},[101,20,116,117,24,27,118,100,119,120,121,34,105,122,104,96,25,103,123,124],"meridian result visualization","result","load","loads","fit","reports","view","performance","health",{"skillNo":126,"name":127,"nameZh":6,"description":128,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":129,"categories":130,"keywords":132,"publishedAt":39},"01M6991WWEVPXCQREVNT95VP64","meridian-model-building","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.","https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fmeridian\u002Fblob\u002Fmain\u002Fskills\u002Fmeridian_model_building\u002FSKILL.md",[131],{"categoryNo":82,"categoryName":83,"taxonomyType":84,"taxonomyTypeDesc":85},[104,20,133,38,134,135,136,25,27,118,28,95,31,120,137,34,105,93,138,139],"meridian model building","data analysis","exploratory data analysis","guides","map","configure","analysis",{"skillNo":141,"name":142,"nameZh":6,"description":143,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":144,"categories":145,"keywords":149,"publishedAt":39},"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",[146],{"categoryNo":147,"categoryName":148,"taxonomyType":89,"taxonomyTypeDesc":90},"01JBSEEDCAT000000000000040","财务",[150,20,151,103,152,153,25,117,27,154,118,100,155,119,34,156,104,101,157,123],"meridian budget optimization","budget optimization","Budget","spending","allocation","budget allocation","find","roi"]