[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill-google-cloud-context-engineering-evaluate":3,"mdc-3picoc-key":35,"related-org-google-cloud-context-engineering-evaluate":1014,"related-repo-google-cloud-context-engineering-evaluate":1201},{"slug":4,"name":4,"fn":5,"description":6,"org":7,"tags":12,"stars":24,"repoUrl":25,"updatedAt":26,"license":27,"forks":28,"topics":29,"repo":30,"sourceUrl":33,"mdContent":34},"context-engineering-evaluate","evaluate ContextSet against golden datasets","Guides the agent to execute an evaluation of a ContextSet against a golden NLQ+SQL dataset using the Evalbench framework.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},"google-cloud","Google Cloud","https:\u002F\u002Fpexgzepcugksgbtrxkhf.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Forg-logos\u002Fgoogle-cloud.png","GoogleCloudPlatform",[13,17,20,23],{"name":14,"slug":15,"type":16},"Evals","evals","tag",{"name":18,"slug":19,"type":16},"Data Analysis","data-analysis",{"name":21,"slug":22,"type":16},"SQL","sql",{"name":9,"slug":8,"type":16},34,"https:\u002F\u002Fgithub.com\u002FGoogleCloudPlatform\u002Fdb-context-enrichment","2026-07-18T05:47:52.594607",null,11,[],{"repoUrl":25,"stars":24,"forks":28,"topics":31,"description":32},[],"A context engineering agent designed to generate, manage, and optimize structured context sets from your database schemas. It bridges the gap between Large Language Models (LLMs) and databases by compiling, evaluating, and maintaining the precise operational context needed for highly accurate natural language-to-SQL query generation.","https:\u002F\u002Fgithub.com\u002FGoogleCloudPlatform\u002Fdb-context-enrichment\u002Ftree\u002FHEAD\u002Fplugin\u002Fskills\u002Fcontext-engineering-evaluate","---\nname: context-engineering-evaluate\ndescription: Guides the agent to execute an evaluation of a ContextSet against a golden NLQ+SQL dataset using the Evalbench framework.\n---\n\n> **Load the `context-engineering-workflow` skill first.** It holds the shared context this phase depends on: workspace layout, state file conventions, phase order, and safety protocol. Do not proceed with this phase without reading it.\n\n# Phase: Evaluation Scoring\n\n## Goal\nRun a structured Evalbench evaluation to score the accuracy of a specific context set against a golden truth dataset, identifying exact query failures.\n\n## Input\n\nBefore beginning the workflow, you explicitly require:\n- A `tools.yaml` file securely located in the workspace root directory containing the target database connection details.\n- A golden evaluation dataset (`golden_dataset_path`), formatted as an absolute system path. The file must be in the **simplified user-facing format**.\n\n  **Simplified User-Facing Dataset Format**:\n  A JSON list of objects, where each object must have the following keys:\n  - `id`: Unique string identifier (e.g., `eval_001`).\n  - `database`: Target database name.\n  - `nlq`: Natural language question.\n  - `golden_sql`: The correct reference SQL query.\n\n  Example:\n  ```json\n  [\n    {\n      \"id\": \"eval_001\",\n      \"database\": \"my_db\",\n      \"nlq\": \"Count users\",\n      \"golden_sql\": \"SELECT COUNT(*) FROM users\"\n    }\n  ]\n  ```\n- The `context_set_id` (the Data Agent's authored context configuration identifier, retrievable by the user directly from the GCP Database Studio console; e.g., `projects\u002F\u003Cproject_id>\u002Flocations\u002F\u003Cregion>\u002FcontextSets\u002F\u003Ccontext_set_name>`).\n\n## Workflow\n\nFollow these steps exactly in order:\n\n1. **Experiment Selection & Memory:**\n   - Scan the local `autoctx\u002Fexperiments\u002F` directory and list the available tuning workflows\u002Fsubfolders to the user.\n   - **If no experiment folders exist** (or the user wants to create a new one without running Bootstrap):\n     - **Ask the user to choose** between 2 paths (do not assume):\n       1. **Bootstrap a basic context**: Guide them to trigger the Bootstrap workflow.\n       2. **Use an existing context**:\n          - > [!IMPORTANT]\n            > Inform the user that if they have an existing context, it must be uploaded to GCP Database Studio to obtain a `context_set_id` for evaluation.\n          - Ask the user for a name for this new experiment folder (similar to how Bootstrap does).\n          - Create the folder under `autoctx\u002Fexperiments\u002F`.\n          - Ask the user to provide the local file path of their existing context.\n          - Record the local file path as the Base Context for this experiment in `autoctx\u002Fstate.md` for long-term memory.\n          - Continue with the evaluation flow below.\n   - Wait for the user to explicitly select an experiment folder to evaluate (or use the newly created one).\n   - Once selected, explicitly record their chosen experiment name into the local `autoctx\u002Fstate.md` file to act as long-term memory so you don't forget it during subsequent evaluations.\n\n2. **Parameter Collection:**\n   - **User Inputs:** Prompt the user ONLY for the `golden_dataset_path` and the `context_set_id` (if they haven't provided them already). Do NOT ask them to explain or verify database configurations.\n   - **Interactive DB Selection:** Read the `autoctx\u002Ftools.yaml` file to list available databases to the user:\n     1. Find all `kind: source` blocks with supported evaluation engines (consult the `generate_evalbench_configs` tool description for the exact list of supported types).\n     2. If there is exactly one *supported* source, inform the user and auto-select it.\n     3. If there are multiple *supported* sources, list their `name` and `type` and let the user select which database to evaluate.\n\n3. **Config Generation (Core Execution):**\n   - Use the `generate_evalbench_configs` MCP tool. This is the **only** way to generate Evalbench configs. Never invent configs from scratch.\n   - If the tool fails, analyze the error and retry with corrected inputs. If it is an internal system error, STOP and inform the user.\n   - Provide the selected `output_dir` (must be `autoctx\u002Fexperiments\u002F\u003Cexperiment_name>\u002F`), `dataset_path`, `context_set_id`, absolute `toolbox_config_path` (e.g. `autoctx\u002Ftools.yaml`), and selected `toolbox_source_name`.\n   - The tool will automatically write all generated configuration files (including `golden_queries.json`) directly to the `eval_configs\u002F` directory inside the chosen `autoctx\u002Fexperiments\u002F\u003Cexperiment_name>\u002F` folder.\n   - You do not need to manually write or extract file contents. Verify that the files have materialized if needed.\n\n4. **Evalbench Run Integration:**\n   - Trigger the `run_shell_command` natively to execute the evaluation from the ROOT of the workspace using the following exact command template:\n     `uvx google-evalbench@1.9.0 --experiment_config=autoctx\u002Fexperiments\u002F\u003Cexperiment_name>\u002Feval_configs\u002Frun_config.yaml`\n   - Check the command outputs to ensure the evaluation reports materialize in the respective `autoctx\u002Fexperiments\u002F\u003Cexperiment_name>\u002Feval_reports\u002F` directory.\n\n## Output\n\nUpon successful completion, the workspace must contain:\n- The generated Evalbench config files successfully written to the `eval_configs\u002F` folder.\n- Evaluating reports built successfully by the external Evalbench runner process.\n\n## Final Summary & Next Steps\n\nConclude by providing a succinct summary to the user:\n- Confirm that the context set has been scored and point out exactly where the final metrics CSV\u002Fresults are located.\n- Share top-level performance summaries.\n- Suggest actionable next steps (e.g., transition to a refinement workflow to hill-climb and improve the metrics based on failed evaluations).\n\n## Templates & Reference\n\nWhen listing sources from `tools.yaml`, ensure you only present `kind: source` records to the user.\nThe tool `generate_evalbench_configs` will find the selected block inside the file and validate its connection parameters deterministically using Python code. You do not need to manually parse or map individual properties such as `host`, `port`, or `database` yourself. If the tool indicates a verification failure for a specific database type, refer to the schema examples inside this directory (e.g., `cloud-sql-postgres.md`) to guide the user on fixing their `tools.yaml` definition.\n",{"data":36,"body":37},{"name":4,"description":6},{"type":38,"children":39},"root",[40,67,74,81,86,92,97,427,433,438,880,886,891,910,916,921,939,945,1008],{"type":41,"tag":42,"props":43,"children":44},"element","blockquote",{},[45],{"type":41,"tag":46,"props":47,"children":48},"p",{},[49,65],{"type":41,"tag":50,"props":51,"children":52},"strong",{},[53,56,63],{"type":54,"value":55},"text","Load the ",{"type":41,"tag":57,"props":58,"children":60},"code",{"className":59},[],[61],{"type":54,"value":62},"context-engineering-workflow",{"type":54,"value":64}," skill first.",{"type":54,"value":66}," It holds the shared context this phase depends on: workspace layout, state file conventions, phase order, and safety protocol. Do not proceed with this phase without reading it.",{"type":41,"tag":68,"props":69,"children":71},"h1",{"id":70},"phase-evaluation-scoring",[72],{"type":54,"value":73},"Phase: Evaluation Scoring",{"type":41,"tag":75,"props":76,"children":78},"h2",{"id":77},"goal",[79],{"type":54,"value":80},"Goal",{"type":41,"tag":46,"props":82,"children":83},{},[84],{"type":54,"value":85},"Run a structured Evalbench evaluation to score the accuracy of a specific context set against a golden truth dataset, identifying exact query failures.",{"type":41,"tag":75,"props":87,"children":89},{"id":88},"input",[90],{"type":54,"value":91},"Input",{"type":41,"tag":46,"props":93,"children":94},{},[95],{"type":54,"value":96},"Before beginning the workflow, you explicitly require:",{"type":41,"tag":98,"props":99,"children":100},"ul",{},[101,115,407],{"type":41,"tag":102,"props":103,"children":104},"li",{},[105,107,113],{"type":54,"value":106},"A ",{"type":41,"tag":57,"props":108,"children":110},{"className":109},[],[111],{"type":54,"value":112},"tools.yaml",{"type":54,"value":114}," file securely located in the workspace root directory containing the target database connection details.",{"type":41,"tag":102,"props":116,"children":117},{},[118,120,126,128,133,135,139,144,146,201,204,206],{"type":54,"value":119},"A golden evaluation dataset (",{"type":41,"tag":57,"props":121,"children":123},{"className":122},[],[124],{"type":54,"value":125},"golden_dataset_path",{"type":54,"value":127},"), formatted as an absolute system path. The file must be in the ",{"type":41,"tag":50,"props":129,"children":130},{},[131],{"type":54,"value":132},"simplified user-facing format",{"type":54,"value":134},".",{"type":41,"tag":136,"props":137,"children":138},"br",{},[],{"type":41,"tag":50,"props":140,"children":141},{},[142],{"type":54,"value":143},"Simplified User-Facing Dataset Format",{"type":54,"value":145},":\nA JSON list of objects, where each object must have the following keys:",{"type":41,"tag":98,"props":147,"children":148},{},[149,168,179,190],{"type":41,"tag":102,"props":150,"children":151},{},[152,158,160,166],{"type":41,"tag":57,"props":153,"children":155},{"className":154},[],[156],{"type":54,"value":157},"id",{"type":54,"value":159},": Unique string identifier (e.g., ",{"type":41,"tag":57,"props":161,"children":163},{"className":162},[],[164],{"type":54,"value":165},"eval_001",{"type":54,"value":167},").",{"type":41,"tag":102,"props":169,"children":170},{},[171,177],{"type":41,"tag":57,"props":172,"children":174},{"className":173},[],[175],{"type":54,"value":176},"database",{"type":54,"value":178},": Target database name.",{"type":41,"tag":102,"props":180,"children":181},{},[182,188],{"type":41,"tag":57,"props":183,"children":185},{"className":184},[],[186],{"type":54,"value":187},"nlq",{"type":54,"value":189},": Natural language question.",{"type":41,"tag":102,"props":191,"children":192},{},[193,199],{"type":41,"tag":57,"props":194,"children":196},{"className":195},[],[197],{"type":54,"value":198},"golden_sql",{"type":54,"value":200},": The correct reference SQL query.",{"type":41,"tag":136,"props":202,"children":203},{},[],{"type":54,"value":205},"Example:",{"type":41,"tag":207,"props":208,"children":213},"pre",{"className":209,"code":210,"language":211,"meta":212,"style":212},"language-json shiki shiki-themes material-theme-lighter material-theme material-theme-palenight","[\n  {\n    \"id\": \"eval_001\",\n    \"database\": \"my_db\",\n    \"nlq\": \"Count users\",\n    \"golden_sql\": \"SELECT COUNT(*) FROM users\"\n  }\n]\n","json","",[214],{"type":41,"tag":57,"props":215,"children":216},{"__ignoreMap":212},[217,229,238,281,318,355,389,398],{"type":41,"tag":218,"props":219,"children":222},"span",{"class":220,"line":221},"line",1,[223],{"type":41,"tag":218,"props":224,"children":226},{"style":225},"--shiki-light:#39ADB5;--shiki-default:#89DDFF;--shiki-dark:#89DDFF",[227],{"type":54,"value":228},"[\n",{"type":41,"tag":218,"props":230,"children":232},{"class":220,"line":231},2,[233],{"type":41,"tag":218,"props":234,"children":235},{"style":225},[236],{"type":54,"value":237},"  {\n",{"type":41,"tag":218,"props":239,"children":241},{"class":220,"line":240},3,[242,247,252,257,262,267,272,276],{"type":41,"tag":218,"props":243,"children":244},{"style":225},[245],{"type":54,"value":246},"    \"",{"type":41,"tag":218,"props":248,"children":250},{"style":249},"--shiki-light:#9C3EDA;--shiki-default:#C792EA;--shiki-dark:#C792EA",[251],{"type":54,"value":157},{"type":41,"tag":218,"props":253,"children":254},{"style":225},[255],{"type":54,"value":256},"\"",{"type":41,"tag":218,"props":258,"children":259},{"style":225},[260],{"type":54,"value":261},":",{"type":41,"tag":218,"props":263,"children":264},{"style":225},[265],{"type":54,"value":266}," \"",{"type":41,"tag":218,"props":268,"children":270},{"style":269},"--shiki-light:#91B859;--shiki-default:#C3E88D;--shiki-dark:#C3E88D",[271],{"type":54,"value":165},{"type":41,"tag":218,"props":273,"children":274},{"style":225},[275],{"type":54,"value":256},{"type":41,"tag":218,"props":277,"children":278},{"style":225},[279],{"type":54,"value":280},",\n",{"type":41,"tag":218,"props":282,"children":284},{"class":220,"line":283},4,[285,289,293,297,301,305,310,314],{"type":41,"tag":218,"props":286,"children":287},{"style":225},[288],{"type":54,"value":246},{"type":41,"tag":218,"props":290,"children":291},{"style":249},[292],{"type":54,"value":176},{"type":41,"tag":218,"props":294,"children":295},{"style":225},[296],{"type":54,"value":256},{"type":41,"tag":218,"props":298,"children":299},{"style":225},[300],{"type":54,"value":261},{"type":41,"tag":218,"props":302,"children":303},{"style":225},[304],{"type":54,"value":266},{"type":41,"tag":218,"props":306,"children":307},{"style":269},[308],{"type":54,"value":309},"my_db",{"type":41,"tag":218,"props":311,"children":312},{"style":225},[313],{"type":54,"value":256},{"type":41,"tag":218,"props":315,"children":316},{"style":225},[317],{"type":54,"value":280},{"type":41,"tag":218,"props":319,"children":321},{"class":220,"line":320},5,[322,326,330,334,338,342,347,351],{"type":41,"tag":218,"props":323,"children":324},{"style":225},[325],{"type":54,"value":246},{"type":41,"tag":218,"props":327,"children":328},{"style":249},[329],{"type":54,"value":187},{"type":41,"tag":218,"props":331,"children":332},{"style":225},[333],{"type":54,"value":256},{"type":41,"tag":218,"props":335,"children":336},{"style":225},[337],{"type":54,"value":261},{"type":41,"tag":218,"props":339,"children":340},{"style":225},[341],{"type":54,"value":266},{"type":41,"tag":218,"props":343,"children":344},{"style":269},[345],{"type":54,"value":346},"Count users",{"type":41,"tag":218,"props":348,"children":349},{"style":225},[350],{"type":54,"value":256},{"type":41,"tag":218,"props":352,"children":353},{"style":225},[354],{"type":54,"value":280},{"type":41,"tag":218,"props":356,"children":358},{"class":220,"line":357},6,[359,363,367,371,375,379,384],{"type":41,"tag":218,"props":360,"children":361},{"style":225},[362],{"type":54,"value":246},{"type":41,"tag":218,"props":364,"children":365},{"style":249},[366],{"type":54,"value":198},{"type":41,"tag":218,"props":368,"children":369},{"style":225},[370],{"type":54,"value":256},{"type":41,"tag":218,"props":372,"children":373},{"style":225},[374],{"type":54,"value":261},{"type":41,"tag":218,"props":376,"children":377},{"style":225},[378],{"type":54,"value":266},{"type":41,"tag":218,"props":380,"children":381},{"style":269},[382],{"type":54,"value":383},"SELECT COUNT(*) FROM users",{"type":41,"tag":218,"props":385,"children":386},{"style":225},[387],{"type":54,"value":388},"\"\n",{"type":41,"tag":218,"props":390,"children":392},{"class":220,"line":391},7,[393],{"type":41,"tag":218,"props":394,"children":395},{"style":225},[396],{"type":54,"value":397},"  }\n",{"type":41,"tag":218,"props":399,"children":401},{"class":220,"line":400},8,[402],{"type":41,"tag":218,"props":403,"children":404},{"style":225},[405],{"type":54,"value":406},"]\n",{"type":41,"tag":102,"props":408,"children":409},{},[410,412,418,420,426],{"type":54,"value":411},"The ",{"type":41,"tag":57,"props":413,"children":415},{"className":414},[],[416],{"type":54,"value":417},"context_set_id",{"type":54,"value":419}," (the Data Agent's authored context configuration identifier, retrievable by the user directly from the GCP Database Studio console; e.g., ",{"type":41,"tag":57,"props":421,"children":423},{"className":422},[],[424],{"type":54,"value":425},"projects\u002F\u003Cproject_id>\u002Flocations\u002F\u003Cregion>\u002FcontextSets\u002F\u003Ccontext_set_name>",{"type":54,"value":167},{"type":41,"tag":75,"props":428,"children":430},{"id":429},"workflow",[431],{"type":54,"value":432},"Workflow",{"type":41,"tag":46,"props":434,"children":435},{},[436],{"type":54,"value":437},"Follow these steps exactly in order:",{"type":41,"tag":439,"props":440,"children":441},"ol",{},[442,594,711,837],{"type":41,"tag":102,"props":443,"children":444},{},[445,450],{"type":41,"tag":50,"props":446,"children":447},{},[448],{"type":54,"value":449},"Experiment Selection & Memory:",{"type":41,"tag":98,"props":451,"children":452},{},[453,466,577,582],{"type":41,"tag":102,"props":454,"children":455},{},[456,458,464],{"type":54,"value":457},"Scan the local ",{"type":41,"tag":57,"props":459,"children":461},{"className":460},[],[462],{"type":54,"value":463},"autoctx\u002Fexperiments\u002F",{"type":54,"value":465}," directory and list the available tuning workflows\u002Fsubfolders to the user.",{"type":41,"tag":102,"props":467,"children":468},{},[469,474,476],{"type":41,"tag":50,"props":470,"children":471},{},[472],{"type":54,"value":473},"If no experiment folders exist",{"type":54,"value":475}," (or the user wants to create a new one without running Bootstrap):\n",{"type":41,"tag":98,"props":477,"children":478},{},[479],{"type":41,"tag":102,"props":480,"children":481},{},[482,487,489],{"type":41,"tag":50,"props":483,"children":484},{},[485],{"type":54,"value":486},"Ask the user to choose",{"type":54,"value":488}," between 2 paths (do not assume):\n",{"type":41,"tag":439,"props":490,"children":491},{},[492,502],{"type":41,"tag":102,"props":493,"children":494},{},[495,500],{"type":41,"tag":50,"props":496,"children":497},{},[498],{"type":54,"value":499},"Bootstrap a basic context",{"type":54,"value":501},": Guide them to trigger the Bootstrap workflow.",{"type":41,"tag":102,"props":503,"children":504},{},[505,510,512],{"type":41,"tag":50,"props":506,"children":507},{},[508],{"type":54,"value":509},"Use an existing context",{"type":54,"value":511},":\n",{"type":41,"tag":98,"props":513,"children":514},{},[515,538,543,554,559,572],{"type":41,"tag":102,"props":516,"children":517},{},[518],{"type":41,"tag":42,"props":519,"children":520},{},[521],{"type":41,"tag":46,"props":522,"children":523},{},[524,529,531,536],{"type":41,"tag":218,"props":525,"children":526},{},[527],{"type":54,"value":528},"!IMPORTANT",{"type":54,"value":530},"\nInform the user that if they have an existing context, it must be uploaded to GCP Database Studio to obtain a ",{"type":41,"tag":57,"props":532,"children":534},{"className":533},[],[535],{"type":54,"value":417},{"type":54,"value":537}," for evaluation.",{"type":41,"tag":102,"props":539,"children":540},{},[541],{"type":54,"value":542},"Ask the user for a name for this new experiment folder (similar to how Bootstrap does).",{"type":41,"tag":102,"props":544,"children":545},{},[546,548,553],{"type":54,"value":547},"Create the folder under ",{"type":41,"tag":57,"props":549,"children":551},{"className":550},[],[552],{"type":54,"value":463},{"type":54,"value":134},{"type":41,"tag":102,"props":555,"children":556},{},[557],{"type":54,"value":558},"Ask the user to provide the local file path of their existing context.",{"type":41,"tag":102,"props":560,"children":561},{},[562,564,570],{"type":54,"value":563},"Record the local file path as the Base Context for this experiment in ",{"type":41,"tag":57,"props":565,"children":567},{"className":566},[],[568],{"type":54,"value":569},"autoctx\u002Fstate.md",{"type":54,"value":571}," for long-term memory.",{"type":41,"tag":102,"props":573,"children":574},{},[575],{"type":54,"value":576},"Continue with the evaluation flow below.",{"type":41,"tag":102,"props":578,"children":579},{},[580],{"type":54,"value":581},"Wait for the user to explicitly select an experiment folder to evaluate (or use the newly created one).",{"type":41,"tag":102,"props":583,"children":584},{},[585,587,592],{"type":54,"value":586},"Once selected, explicitly record their chosen experiment name into the local ",{"type":41,"tag":57,"props":588,"children":590},{"className":589},[],[591],{"type":54,"value":569},{"type":54,"value":593}," file to act as long-term memory so you don't forget it during subsequent evaluations.",{"type":41,"tag":102,"props":595,"children":596},{},[597,602],{"type":41,"tag":50,"props":598,"children":599},{},[600],{"type":54,"value":601},"Parameter Collection:",{"type":41,"tag":98,"props":603,"children":604},{},[605,629],{"type":41,"tag":102,"props":606,"children":607},{},[608,613,615,620,622,627],{"type":41,"tag":50,"props":609,"children":610},{},[611],{"type":54,"value":612},"User Inputs:",{"type":54,"value":614}," Prompt the user ONLY for the ",{"type":41,"tag":57,"props":616,"children":618},{"className":617},[],[619],{"type":54,"value":125},{"type":54,"value":621}," and the ",{"type":41,"tag":57,"props":623,"children":625},{"className":624},[],[626],{"type":54,"value":417},{"type":54,"value":628}," (if they haven't provided them already). Do NOT ask them to explain or verify database configurations.",{"type":41,"tag":102,"props":630,"children":631},{},[632,637,639,645,647],{"type":41,"tag":50,"props":633,"children":634},{},[635],{"type":54,"value":636},"Interactive DB Selection:",{"type":54,"value":638}," Read the ",{"type":41,"tag":57,"props":640,"children":642},{"className":641},[],[643],{"type":54,"value":644},"autoctx\u002Ftools.yaml",{"type":54,"value":646}," file to list available databases to the user:\n",{"type":41,"tag":439,"props":648,"children":649},{},[650,671,684],{"type":41,"tag":102,"props":651,"children":652},{},[653,655,661,663,669],{"type":54,"value":654},"Find all ",{"type":41,"tag":57,"props":656,"children":658},{"className":657},[],[659],{"type":54,"value":660},"kind: source",{"type":54,"value":662}," blocks with supported evaluation engines (consult the ",{"type":41,"tag":57,"props":664,"children":666},{"className":665},[],[667],{"type":54,"value":668},"generate_evalbench_configs",{"type":54,"value":670}," tool description for the exact list of supported types).",{"type":41,"tag":102,"props":672,"children":673},{},[674,676,682],{"type":54,"value":675},"If there is exactly one ",{"type":41,"tag":677,"props":678,"children":679},"em",{},[680],{"type":54,"value":681},"supported",{"type":54,"value":683}," source, inform the user and auto-select it.",{"type":41,"tag":102,"props":685,"children":686},{},[687,689,693,695,701,703,709],{"type":54,"value":688},"If there are multiple ",{"type":41,"tag":677,"props":690,"children":691},{},[692],{"type":54,"value":681},{"type":54,"value":694}," sources, list their ",{"type":41,"tag":57,"props":696,"children":698},{"className":697},[],[699],{"type":54,"value":700},"name",{"type":54,"value":702}," and ",{"type":41,"tag":57,"props":704,"children":706},{"className":705},[],[707],{"type":54,"value":708},"type",{"type":54,"value":710}," and let the user select which database to evaluate.",{"type":41,"tag":102,"props":712,"children":713},{},[714,719],{"type":41,"tag":50,"props":715,"children":716},{},[717],{"type":54,"value":718},"Config Generation (Core Execution):",{"type":41,"tag":98,"props":720,"children":721},{},[722,741,746,804,832],{"type":41,"tag":102,"props":723,"children":724},{},[725,727,732,734,739],{"type":54,"value":726},"Use the ",{"type":41,"tag":57,"props":728,"children":730},{"className":729},[],[731],{"type":54,"value":668},{"type":54,"value":733}," MCP tool. This is the ",{"type":41,"tag":50,"props":735,"children":736},{},[737],{"type":54,"value":738},"only",{"type":54,"value":740}," way to generate Evalbench configs. Never invent configs from scratch.",{"type":41,"tag":102,"props":742,"children":743},{},[744],{"type":54,"value":745},"If the tool fails, analyze the error and retry with corrected inputs. If it is an internal system error, STOP and inform the user.",{"type":41,"tag":102,"props":747,"children":748},{},[749,751,757,759,765,767,773,775,780,782,788,790,795,797,803],{"type":54,"value":750},"Provide the selected ",{"type":41,"tag":57,"props":752,"children":754},{"className":753},[],[755],{"type":54,"value":756},"output_dir",{"type":54,"value":758}," (must be ",{"type":41,"tag":57,"props":760,"children":762},{"className":761},[],[763],{"type":54,"value":764},"autoctx\u002Fexperiments\u002F\u003Cexperiment_name>\u002F",{"type":54,"value":766},"), ",{"type":41,"tag":57,"props":768,"children":770},{"className":769},[],[771],{"type":54,"value":772},"dataset_path",{"type":54,"value":774},", ",{"type":41,"tag":57,"props":776,"children":778},{"className":777},[],[779],{"type":54,"value":417},{"type":54,"value":781},", absolute ",{"type":41,"tag":57,"props":783,"children":785},{"className":784},[],[786],{"type":54,"value":787},"toolbox_config_path",{"type":54,"value":789}," (e.g. ",{"type":41,"tag":57,"props":791,"children":793},{"className":792},[],[794],{"type":54,"value":644},{"type":54,"value":796},"), and selected ",{"type":41,"tag":57,"props":798,"children":800},{"className":799},[],[801],{"type":54,"value":802},"toolbox_source_name",{"type":54,"value":134},{"type":41,"tag":102,"props":805,"children":806},{},[807,809,815,817,823,825,830],{"type":54,"value":808},"The tool will automatically write all generated configuration files (including ",{"type":41,"tag":57,"props":810,"children":812},{"className":811},[],[813],{"type":54,"value":814},"golden_queries.json",{"type":54,"value":816},") directly to the ",{"type":41,"tag":57,"props":818,"children":820},{"className":819},[],[821],{"type":54,"value":822},"eval_configs\u002F",{"type":54,"value":824}," directory inside the chosen ",{"type":41,"tag":57,"props":826,"children":828},{"className":827},[],[829],{"type":54,"value":764},{"type":54,"value":831}," folder.",{"type":41,"tag":102,"props":833,"children":834},{},[835],{"type":54,"value":836},"You do not need to manually write or extract file contents. Verify that the files have materialized if needed.",{"type":41,"tag":102,"props":838,"children":839},{},[840,845],{"type":41,"tag":50,"props":841,"children":842},{},[843],{"type":54,"value":844},"Evalbench Run Integration:",{"type":41,"tag":98,"props":846,"children":847},{},[848,867],{"type":41,"tag":102,"props":849,"children":850},{},[851,853,859,861],{"type":54,"value":852},"Trigger the ",{"type":41,"tag":57,"props":854,"children":856},{"className":855},[],[857],{"type":54,"value":858},"run_shell_command",{"type":54,"value":860}," natively to execute the evaluation from the ROOT of the workspace using the following exact command template:\n",{"type":41,"tag":57,"props":862,"children":864},{"className":863},[],[865],{"type":54,"value":866},"uvx google-evalbench@1.9.0 --experiment_config=autoctx\u002Fexperiments\u002F\u003Cexperiment_name>\u002Feval_configs\u002Frun_config.yaml",{"type":41,"tag":102,"props":868,"children":869},{},[870,872,878],{"type":54,"value":871},"Check the command outputs to ensure the evaluation reports materialize in the respective ",{"type":41,"tag":57,"props":873,"children":875},{"className":874},[],[876],{"type":54,"value":877},"autoctx\u002Fexperiments\u002F\u003Cexperiment_name>\u002Feval_reports\u002F",{"type":54,"value":879}," directory.",{"type":41,"tag":75,"props":881,"children":883},{"id":882},"output",[884],{"type":54,"value":885},"Output",{"type":41,"tag":46,"props":887,"children":888},{},[889],{"type":54,"value":890},"Upon successful completion, the workspace must contain:",{"type":41,"tag":98,"props":892,"children":893},{},[894,905],{"type":41,"tag":102,"props":895,"children":896},{},[897,899,904],{"type":54,"value":898},"The generated Evalbench config files successfully written to the ",{"type":41,"tag":57,"props":900,"children":902},{"className":901},[],[903],{"type":54,"value":822},{"type":54,"value":831},{"type":41,"tag":102,"props":906,"children":907},{},[908],{"type":54,"value":909},"Evaluating reports built successfully by the external Evalbench runner process.",{"type":41,"tag":75,"props":911,"children":913},{"id":912},"final-summary-next-steps",[914],{"type":54,"value":915},"Final Summary & Next Steps",{"type":41,"tag":46,"props":917,"children":918},{},[919],{"type":54,"value":920},"Conclude by providing a succinct summary to the user:",{"type":41,"tag":98,"props":922,"children":923},{},[924,929,934],{"type":41,"tag":102,"props":925,"children":926},{},[927],{"type":54,"value":928},"Confirm that the context set has been scored and point out exactly where the final metrics CSV\u002Fresults are located.",{"type":41,"tag":102,"props":930,"children":931},{},[932],{"type":54,"value":933},"Share top-level performance summaries.",{"type":41,"tag":102,"props":935,"children":936},{},[937],{"type":54,"value":938},"Suggest actionable next steps (e.g., transition to a refinement workflow to hill-climb and improve the metrics based on failed evaluations).",{"type":41,"tag":75,"props":940,"children":942},{"id":941},"templates-reference",[943],{"type":54,"value":944},"Templates & Reference",{"type":41,"tag":46,"props":946,"children":947},{},[948,950,955,957,962,964,969,971,977,978,984,986,991,993,999,1001,1006],{"type":54,"value":949},"When listing sources from ",{"type":41,"tag":57,"props":951,"children":953},{"className":952},[],[954],{"type":54,"value":112},{"type":54,"value":956},", ensure you only present ",{"type":41,"tag":57,"props":958,"children":960},{"className":959},[],[961],{"type":54,"value":660},{"type":54,"value":963}," records to the user.\nThe tool ",{"type":41,"tag":57,"props":965,"children":967},{"className":966},[],[968],{"type":54,"value":668},{"type":54,"value":970}," will find the selected block inside the file and validate its connection parameters deterministically using Python code. You do not need to manually parse or map individual properties such as ",{"type":41,"tag":57,"props":972,"children":974},{"className":973},[],[975],{"type":54,"value":976},"host",{"type":54,"value":774},{"type":41,"tag":57,"props":979,"children":981},{"className":980},[],[982],{"type":54,"value":983},"port",{"type":54,"value":985},", or ",{"type":41,"tag":57,"props":987,"children":989},{"className":988},[],[990],{"type":54,"value":176},{"type":54,"value":992}," yourself. If the tool indicates a verification failure for a specific database type, refer to the schema examples inside this directory (e.g., ",{"type":41,"tag":57,"props":994,"children":996},{"className":995},[],[997],{"type":54,"value":998},"cloud-sql-postgres.md",{"type":54,"value":1000},") to guide the user on fixing their ",{"type":41,"tag":57,"props":1002,"children":1004},{"className":1003},[],[1005],{"type":54,"value":112},{"type":54,"value":1007}," definition.",{"type":41,"tag":1009,"props":1010,"children":1011},"style",{},[1012],{"type":54,"value":1013},"html .light .shiki span {color: var(--shiki-light);background: var(--shiki-light-bg);font-style: var(--shiki-light-font-style);font-weight: var(--shiki-light-font-weight);text-decoration: var(--shiki-light-text-decoration);}html.light .shiki span {color: var(--shiki-light);background: var(--shiki-light-bg);font-style: var(--shiki-light-font-style);font-weight: var(--shiki-light-font-weight);text-decoration: var(--shiki-light-text-decoration);}html .default .shiki span {color: var(--shiki-default);background: var(--shiki-default-bg);font-style: var(--shiki-default-font-style);font-weight: var(--shiki-default-font-weight);text-decoration: var(--shiki-default-text-decoration);}html .shiki span {color: var(--shiki-default);background: var(--shiki-default-bg);font-style: var(--shiki-default-font-style);font-weight: var(--shiki-default-font-weight);text-decoration: var(--shiki-default-text-decoration);}html .dark .shiki span {color: var(--shiki-dark);background: var(--shiki-dark-bg);font-style: var(--shiki-dark-font-style);font-weight: var(--shiki-dark-font-weight);text-decoration: var(--shiki-dark-text-decoration);}html.dark .shiki span {color: var(--shiki-dark);background: var(--shiki-dark-bg);font-style: var(--shiki-dark-font-style);font-weight: var(--shiki-dark-font-weight);text-decoration: var(--shiki-dark-text-decoration);}",{"items":1015,"total":1200},[1016,1034,1048,1070,1084,1095,1109,1126,1143,1156,1172,1182],{"slug":1017,"name":1017,"fn":1018,"description":1019,"org":1020,"tags":1021,"stars":1031,"repoUrl":1032,"updatedAt":1033},"kb-search","search and extract local knowledge base documents","Allows listing, searching and extracting information from local knowledge base documents for information about tables\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1022,1025,1028],{"name":1023,"slug":1024,"type":16},"Documentation","documentation",{"name":1026,"slug":1027,"type":16},"Knowledge Base","knowledge-base",{"name":1029,"slug":1030,"type":16},"Search","search",6749,"https:\u002F\u002Fgithub.com\u002FGoogleCloudPlatform\u002Fknowledge-catalog","2026-07-12T07:38:52.157375",{"slug":1035,"name":1036,"fn":1037,"description":1038,"org":1039,"tags":1040,"stars":1031,"repoUrl":1032,"updatedAt":1047},"knowledgecatalogdiscoveryagent","knowledge_catalog_discovery_agent","search and rank Knowledge Catalog data entries","Analyzes user queries, extracts relevant predicates, and utilizes Knowledge Catalog Search to find and rank the most relevant data entries. Engages with the user throughout the process.\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1041,1042,1043,1046],{"name":18,"slug":19,"type":16},{"name":9,"slug":8,"type":16},{"name":1044,"slug":1045,"type":16},"Knowledge Management","knowledge-management",{"name":1029,"slug":1030,"type":16},"2026-07-12T07:38:22.196851",{"slug":1049,"name":1049,"fn":1050,"description":1051,"org":1052,"tags":1053,"stars":1067,"repoUrl":1068,"updatedAt":1069},"contributing","contribute to Cloud Foundation Fabric","End-to-end workflow for contributing to Cloud Foundation Fabric: triaging GitHub issues, proactive feature development, validating with tests and Policy Troubleshooter, and submitting sanitized Pull Requests. Use when addressing a Fabric GitHub issue, developing a module or FAST stage change, or preparing a branch for a pull request.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1054,1057,1060,1063,1064],{"name":1055,"slug":1056,"type":16},"Automation","automation",{"name":1058,"slug":1059,"type":16},"Engineering","engineering",{"name":1061,"slug":1062,"type":16},"GitHub","github",{"name":9,"slug":8,"type":16},{"name":1065,"slug":1066,"type":16},"Pull Requests","pull-requests",2062,"https:\u002F\u002Fgithub.com\u002FGoogleCloudPlatform\u002Fcloud-foundation-fabric","2026-07-31T06:23:36.935005",{"slug":1071,"name":1071,"fn":1072,"description":1073,"org":1074,"tags":1075,"stars":1067,"repoUrl":1068,"updatedAt":1083},"fabric-builder","generate Terraform code for Google Cloud","Generates idiomatic Cloud Foundation Fabric (CFF) Terraform code using CFF modules. Use when users ask to create GCP resources, use Fabric modules, or generate Terraform code for Google Cloud.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1076,1077,1080],{"name":9,"slug":8,"type":16},{"name":1078,"slug":1079,"type":16},"Infrastructure as Code","infrastructure-as-code",{"name":1081,"slug":1082,"type":16},"Terraform","terraform","2026-07-12T07:38:23.514555",{"slug":1085,"name":1085,"fn":1086,"description":1087,"org":1088,"tags":1089,"stars":1067,"repoUrl":1068,"updatedAt":1094},"fast-0-org-setup-prereqs","prepare prerequisites for FAST 0-org-setup","Guides the user step-by-step through the prerequisites for the FAST 0-org-setup stage, supporting both Standard GCP and Google Cloud Dedicated (GCD) environments. Use when a user asks to prepare or run prerequisites for 0-org-setup or bootstrap the FAST landing zone.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1090,1091],{"name":9,"slug":8,"type":16},{"name":1092,"slug":1093,"type":16},"Operations","operations","2026-07-12T07:38:28.127148",{"slug":1096,"name":1096,"fn":1097,"description":1098,"org":1099,"tags":1100,"stars":1106,"repoUrl":1107,"updatedAt":1108},"agent-aware-cli","design agent-aware command-line interfaces","Guide for designing and implementing command-line interfaces (CLIs) that are equally usable by human developers and automated coding agents. Use when the user wants to build a CLI, apply CLI best practices, or use Go with Cobra and Viper.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1101,1104,1105],{"name":1102,"slug":1103,"type":16},"CLI","cli",{"name":1058,"slug":1059,"type":16},{"name":9,"slug":8,"type":16},1150,"https:\u002F\u002Fgithub.com\u002FGoogleCloudPlatform\u002Fvertex-ai-creative-studio","2026-07-12T07:39:08.41406",{"slug":1110,"name":1110,"fn":1111,"description":1112,"org":1113,"tags":1114,"stars":1106,"repoUrl":1107,"updatedAt":1125},"build-mcp-genmedia","build and configure GenAI MCP servers","Builds the mcp-genmedia Go MCP servers (nanobanana, veo, lyria, gemini-multimodal, chirp3-hd, avtool) from source and wires them into settings.json. Use this skill whenever the MCP tools are missing or broken — typically at the start of a new session, after a container restart, or when \u002Ftmp has been wiped. The prebuilt binaries in \u002Fworkspace\u002F.local\u002Fbin\u002F have no exec bit and live on a noexec mount; this skill compiles fresh executables into \u002Ftmp\u002Fbin\u002F where execution is allowed.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1115,1118,1119,1122],{"name":1116,"slug":1117,"type":16},"API Development","api-development",{"name":9,"slug":8,"type":16},{"name":1120,"slug":1121,"type":16},"LLM","llm",{"name":1123,"slug":1124,"type":16},"MCP","mcp","2026-07-12T07:39:10.911302",{"slug":1127,"name":1127,"fn":1128,"description":1129,"org":1130,"tags":1131,"stars":1106,"repoUrl":1107,"updatedAt":1142},"genmedia-audio-engineer","synthesize and mix audio content","Expert in audio synthesis, music generation, and mixing. Use when creating podcasts, background scores, or multi-track audio layering using mcp-chirp3-go, mcp-lyria-go, mcp-gemini-go, mcp-nanobanana-go, and mcp-avtool-go.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1132,1135,1138,1139],{"name":1133,"slug":1134,"type":16},"Audio","audio",{"name":1136,"slug":1137,"type":16},"Creative","creative",{"name":9,"slug":8,"type":16},{"name":1140,"slug":1141,"type":16},"Vertex AI","vertex-ai","2026-07-12T07:39:16.623879",{"slug":1144,"name":1144,"fn":1145,"description":1146,"org":1147,"tags":1148,"stars":1106,"repoUrl":1107,"updatedAt":1155},"genmedia-image-artist","generate and edit AI images","Expert in AI image generation and editing. Use when the user needs high-quality textures, character-consistent visuals, or image-to-image editing using mcp-nanobanana-go.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1149,1150,1151,1154],{"name":1136,"slug":1137,"type":16},{"name":9,"slug":8,"type":16},{"name":1152,"slug":1153,"type":16},"Image Generation","image-generation",{"name":1140,"slug":1141,"type":16},"2026-07-12T07:39:15.372822",{"slug":1157,"name":1157,"fn":1158,"description":1159,"org":1160,"tags":1161,"stars":1106,"repoUrl":1107,"updatedAt":1171},"genmedia-producer","produce multi-step media content","Expert media production assistant. Use when requested to help with storyboarding, podcast creation, audio assembly, or complex multi-step media workflows using the GenMedia MCP servers (Veo, Lyria, Gemini TTS, NanoBanana).",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1162,1163,1164,1165,1168],{"name":1133,"slug":1134,"type":16},{"name":1136,"slug":1137,"type":16},{"name":9,"slug":8,"type":16},{"name":1166,"slug":1167,"type":16},"Media","media",{"name":1169,"slug":1170,"type":16},"Video","video","2026-07-12T07:39:09.672849",{"slug":1173,"name":1173,"fn":1174,"description":1175,"org":1176,"tags":1177,"stars":1106,"repoUrl":1107,"updatedAt":1181},"genmedia-video-editor","edit and compose video content","Expert in video composition, editing, and format conversion. Use when the user wants to generate high-quality video, overlay images on video, concatenate clips, create GIFs, or sync audio to video using mcp-avtool-go and mcp-veo-go.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1178,1179,1180],{"name":1136,"slug":1137,"type":16},{"name":9,"slug":8,"type":16},{"name":1169,"slug":1170,"type":16},"2026-07-12T07:39:13.749081",{"slug":1183,"name":1183,"fn":1184,"description":1185,"org":1186,"tags":1187,"stars":1106,"repoUrl":1107,"updatedAt":1199},"genmedia-voice-director","generate expressive text-to-speech with Gemini","Expert in casting, directing, and generating expressive text-to-speech using Gemini TTS. Use this when the user needs virtual voice actor personas, expressive speech generation, or multiple variations of a voiceover (like \"take 3 on the bounce\").",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1188,1189,1190,1193,1196],{"name":1133,"slug":1134,"type":16},{"name":1136,"slug":1137,"type":16},{"name":1191,"slug":1192,"type":16},"Gemini","gemini",{"name":1194,"slug":1195,"type":16},"Speech","speech",{"name":1197,"slug":1198,"type":16},"Text-to-Speech","text-to-speech","2026-07-12T07:39:17.86673",80,{"items":1202,"total":391},[1203,1219,1230,1237,1250,1263,1275],{"slug":1204,"name":1204,"fn":1205,"description":1206,"org":1207,"tags":1208,"stars":24,"repoUrl":25,"updatedAt":1218},"context-engineering-bootstrap","bootstrap initial ContextSet for database enrichment","Guides the agent to bootstrap an initial ContextSet (templates, facets, and value searches) by deducing key information from the database schema and generating a ContextSet file.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1209,1212,1215,1217],{"name":1210,"slug":1211,"type":16},"Context","context",{"name":1213,"slug":1214,"type":16},"Data Modeling","data-modeling",{"name":1216,"slug":176,"type":16},"Database",{"name":9,"slug":8,"type":16},"2026-07-18T05:47:56.797107",{"slug":1220,"name":1220,"fn":1221,"description":1222,"org":1223,"tags":1224,"stars":24,"repoUrl":25,"updatedAt":1229},"context-engineering-dataset-generation","generate golden SQL evaluation datasets","Generate or expand a golden evaluation dataset of SQL\u002FQuestion (NLQ+SQL) pairs for evaluating NL-to-SQL translation accuracy on a target database.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1225,1226,1227,1228],{"name":18,"slug":19,"type":16},{"name":14,"slug":15,"type":16},{"name":9,"slug":8,"type":16},{"name":21,"slug":22,"type":16},"2026-07-18T05:47:55.878915",{"slug":4,"name":4,"fn":5,"description":6,"org":1231,"tags":1232,"stars":24,"repoUrl":25,"updatedAt":26},{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1233,1234,1235,1236],{"name":18,"slug":19,"type":16},{"name":14,"slug":15,"type":16},{"name":9,"slug":8,"type":16},{"name":21,"slug":22,"type":16},{"slug":1238,"name":1238,"fn":1239,"description":1240,"org":1241,"tags":1242,"stars":24,"repoUrl":25,"updatedAt":1249},"context-engineering-hillclimb","improve ContextSet via hill-climbing iterations","Guides the agent to perform hill-climbing iterations to improve a ContextSet based on Evalbench evaluation results.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1243,1244,1245,1246],{"name":18,"slug":19,"type":16},{"name":14,"slug":15,"type":16},{"name":9,"slug":8,"type":16},{"name":1247,"slug":1248,"type":16},"Optimization","optimization","2026-07-18T05:47:54.951176",{"slug":1251,"name":1251,"fn":1252,"description":1253,"org":1254,"tags":1255,"stars":24,"repoUrl":25,"updatedAt":1262},"context-engineering-init","initialize context engineering and database connections","Orchestrates the initialization workflow for context engineering, and provides helper workflow for setting up database connections by creating or updating tools.yaml configurations.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1256,1259,1260,1261],{"name":1257,"slug":1258,"type":16},"Configuration","configuration",{"name":1216,"slug":176,"type":16},{"name":1058,"slug":1059,"type":16},{"name":9,"slug":8,"type":16},"2026-07-18T05:47:59.493278",{"slug":62,"name":62,"fn":1264,"description":1265,"org":1266,"tags":1267,"stars":24,"repoUrl":25,"updatedAt":1274},"optimize context sets for Gemini data agents","Context engineering for Gemini Data Analytics API's data agent developer platform tools. Generates, evaluates, and iteratively optimizes a ContextSet (Templates, Facets, Value Searches) to maximize Natural-Language-to-SQL translation accuracy. Use this skill to run the automated setup, NL-SQL pair evaluation dataset generation and expansion, bootstrapping, scoring, and optimization pipeline. For manual authoring standards and schema syntax rules, see the context-generation-guide skill.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1268,1271,1272,1273],{"name":1269,"slug":1270,"type":16},"Agent Context","agent-context",{"name":18,"slug":19,"type":16},{"name":1191,"slug":1192,"type":16},{"name":9,"slug":8,"type":16},"2026-07-18T05:13:44.430929",{"slug":1276,"name":1276,"fn":1277,"description":1278,"org":1279,"tags":1280,"stars":24,"repoUrl":25,"updatedAt":1285},"context-generation-guide","generate context items for database enrichment","Guidelines and best practices for generating context items (Templates, Facets, Value Searches). Use this skill whenever the user asks to create, author, or generate context for database enrichment, or asks for examples and instructions on how to write templates, facets, or value searches. It helps bridge the gap between LLMs and structured databases. For running the automated generation, evaluation, and tuning lifecycle, see the context-engineering-workflow skill.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[1281,1282,1283,1284],{"name":1269,"slug":1270,"type":16},{"name":1216,"slug":176,"type":16},{"name":1023,"slug":1024,"type":16},{"name":9,"slug":8,"type":16},"2026-07-18T05:13:43.403052"]