[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill-google-cloud-eval-breakdown":3,"mdc--hcn4ml-key":44,"related-repo-google-cloud-eval-breakdown":465,"related-org-google-cloud-eval-breakdown":500},{"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":39,"sourceUrl":42,"mdContent":43},"eval-breakdown","diagnose agent evaluation benchmark results","Performs an exhaustive, question-by-question narrative diagnostic breakdown of an agent-eval benchmark run by analyzing question_answer_log.md, eval_summary.json, and raw trajectory traces. Use when diagnosing low score causes, investigating the Memory Reuse vs. Traceability rubric clash, performing pre-release failure audits, or examining judge reasoning across individual scenarios. Don't use for running the benchmark CLI pipeline itself (use agent-eval) or automated genetic prompt tuning (use google-agents-cli-eval).",{"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",[13,17,20,21],{"name":14,"slug":15,"type":16},"Evals","evals","tag",{"name":18,"slug":19,"type":16},"Code Analysis","code-analysis",{"name":9,"slug":8,"type":16},{"name":22,"slug":23,"type":16},"Debugging","debugging",3059,"https:\u002F\u002Fgithub.com\u002FGoogleCloudPlatform\u002Fprofessional-services","2026-08-13T05:04:18.336276",null,1465,[30,31,32,33,34,35,36,37,38],"bigquery","examples","gke","google-cloud-compute","google-cloud-dataflow","google-cloud-ml","google-cloud-platform","solutions","tools",{"repoUrl":25,"stars":24,"forks":28,"topics":40,"description":41},[30,31,32,33,34,35,36,37,38],"Common solutions and tools developed by Google Cloud's Professional Services team. This repository and its contents are not an officially supported Google product.","https:\u002F\u002Fgithub.com\u002FGoogleCloudPlatform\u002Fprofessional-services\u002Ftree\u002FHEAD\u002Ftools\u002Fagent-eval\u002Fskills\u002Feval-breakdown","---\nname: eval-breakdown\ndescription: >-\n  Performs an exhaustive, question-by-question narrative diagnostic breakdown of an agent-eval benchmark run by analyzing question_answer_log.md, eval_summary.json, and raw trajectory traces. Use when diagnosing low score causes, investigating the Memory Reuse vs. Traceability rubric clash, performing pre-release failure audits, or examining judge reasoning across individual scenarios. Don't use for running the benchmark CLI pipeline itself (use agent-eval) or automated genetic prompt tuning (use google-agents-cli-eval).\nmetadata:\n  author: Google Cloud Professional Services\n  license: Apache-2.0\n  version: 1.0.0\n  requires:\n    bins:\n      - agent-eval\n---\n\n# Eval Breakdown: Question-by-Question Diagnostic Analysis\n\nThis skill guides an agent through conducting a rigorous, evidence-grounded, question-by-question narrative diagnostic audit of an `agent-eval` benchmark run.\n\n---\n\n## 1. Input Sources & Directory Anatomy\n\nAn `agent-eval` results folder (`tests\u002Feval\u002Fresults\u002F{run_id}\u002F` or `gs:\u002F\u002F\u003Cbucket>\u002Fruns\u002F{run_id}\u002F`) contains three primary diagnostic artifacts:\n1. `eval_summary.json`: Aggregated metrics, per-question score distributions, and AutoRater judge verdicts.\n2. `question_answer_log.md`: Multi-turn conversational transcripts including user prompts, agent responses, tool calls, and tool outputs.\n3. `gemini_analysis.md`: Automated executive diagnosis and loss cluster summary.\n\n---\n\n## 2. Step-by-Step Diagnostic Workflow\n\n### Step 1: Generate the Baseline Matrix Table\nExtract individual question scores and judge reasoning using the bundled script:\n\n```bash\npython3 tools\u002Fagent-eval\u002Fskills\u002Fagent-eval\u002Fscripts\u002Fparse_eval_summary.py \\\n  --summary-path tests\u002Feval\u002Fresults\u002F{run_id}\u002Feval_summary.json\n```\n\n### Step 2: Perform the Dialogue Audit Protocol\nInspect `question_answer_log.md` for each scenario, analyzing across:\n1. **Turn-by-Turn User Intent**: What the user requested in Turn 1, Turn 2, etc.\n2. **Tool Execution & Traceability**: Did the agent emit SQL\u002FAPI tool calls on each turn where factual assertions were made, or did it answer from empirical dialogue memory?\n3. **The Memory Reuse vs. Traceability Rubric Clash**:\n   * *Diagnostic Pattern*: If `tool_use_quality` drops to `0.00` on Turn 2 follow-ups, verify if the agent answered correctly from dialogue memory without re-querying the backend. If the rubric expects a tool call on every turn, calibrate the rubric or prompt instructions accordingly.\n4. **Judge Explanation Grounding**: Cross-reference any rubric score $\u003C 1.00$ against the exact model response to verify whether the deduction was a genuine agent failure (Tier 1) or an overly strict judge rubric (Tier 2).\n\n---\n\n## 3. Executive Output Template\n\nConclude the audit with a structured summary table:\n\n| Q# | Scenario ID | Metrics Summary | Root-Cause Finding |\n| :--- | :--- | :--- | :--- |\n| `Q1` | `TC-001` | `tool_use`=1.0, `business_logic`=1.0 | ✅ PASS (Strict parameter adherence) |\n| `Q2` | `TC-002` | `tool_use`=0.0, `business_logic`=1.0 | ⚠️ Memory reuse on Turn 2; rubric expected re-query |\n",{"data":45,"body":53},{"name":4,"description":6,"metadata":46},{"author":47,"license":48,"version":49,"requires":50},"Google Cloud Professional Services","Apache-2.0","1.0.0",{"bins":51},[52],"agent-eval",{"type":54,"children":55},"root",[56,65,79,83,90,118,156,159,165,172,177,226,232,244,319,322,328,333,459],{"type":57,"tag":58,"props":59,"children":61},"element","h1",{"id":60},"eval-breakdown-question-by-question-diagnostic-analysis",[62],{"type":63,"value":64},"text","Eval Breakdown: Question-by-Question Diagnostic Analysis",{"type":57,"tag":66,"props":67,"children":68},"p",{},[69,71,77],{"type":63,"value":70},"This skill guides an agent through conducting a rigorous, evidence-grounded, question-by-question narrative diagnostic audit of an ",{"type":57,"tag":72,"props":73,"children":75},"code",{"className":74},[],[76],{"type":63,"value":52},{"type":63,"value":78}," benchmark run.",{"type":57,"tag":80,"props":81,"children":82},"hr",{},[],{"type":57,"tag":84,"props":85,"children":87},"h2",{"id":86},"_1-input-sources-directory-anatomy",[88],{"type":63,"value":89},"1. Input Sources & Directory Anatomy",{"type":57,"tag":66,"props":91,"children":92},{},[93,95,100,102,108,110,116],{"type":63,"value":94},"An ",{"type":57,"tag":72,"props":96,"children":98},{"className":97},[],[99],{"type":63,"value":52},{"type":63,"value":101}," results folder (",{"type":57,"tag":72,"props":103,"children":105},{"className":104},[],[106],{"type":63,"value":107},"tests\u002Feval\u002Fresults\u002F{run_id}\u002F",{"type":63,"value":109}," or ",{"type":57,"tag":72,"props":111,"children":113},{"className":112},[],[114],{"type":63,"value":115},"gs:\u002F\u002F\u003Cbucket>\u002Fruns\u002F{run_id}\u002F",{"type":63,"value":117},") contains three primary diagnostic artifacts:",{"type":57,"tag":119,"props":120,"children":121},"ol",{},[122,134,145],{"type":57,"tag":123,"props":124,"children":125},"li",{},[126,132],{"type":57,"tag":72,"props":127,"children":129},{"className":128},[],[130],{"type":63,"value":131},"eval_summary.json",{"type":63,"value":133},": Aggregated metrics, per-question score distributions, and AutoRater judge verdicts.",{"type":57,"tag":123,"props":135,"children":136},{},[137,143],{"type":57,"tag":72,"props":138,"children":140},{"className":139},[],[141],{"type":63,"value":142},"question_answer_log.md",{"type":63,"value":144},": Multi-turn conversational transcripts including user prompts, agent responses, tool calls, and tool outputs.",{"type":57,"tag":123,"props":146,"children":147},{},[148,154],{"type":57,"tag":72,"props":149,"children":151},{"className":150},[],[152],{"type":63,"value":153},"gemini_analysis.md",{"type":63,"value":155},": Automated executive diagnosis and loss cluster summary.",{"type":57,"tag":80,"props":157,"children":158},{},[],{"type":57,"tag":84,"props":160,"children":162},{"id":161},"_2-step-by-step-diagnostic-workflow",[163],{"type":63,"value":164},"2. Step-by-Step Diagnostic Workflow",{"type":57,"tag":166,"props":167,"children":169},"h3",{"id":168},"step-1-generate-the-baseline-matrix-table",[170],{"type":63,"value":171},"Step 1: Generate the Baseline Matrix Table",{"type":57,"tag":66,"props":173,"children":174},{},[175],{"type":63,"value":176},"Extract individual question scores and judge reasoning using the bundled script:",{"type":57,"tag":178,"props":179,"children":184},"pre",{"className":180,"code":181,"language":182,"meta":183,"style":183},"language-bash shiki shiki-themes material-theme-lighter material-theme material-theme-palenight","python3 tools\u002Fagent-eval\u002Fskills\u002Fagent-eval\u002Fscripts\u002Fparse_eval_summary.py \\\n  --summary-path tests\u002Feval\u002Fresults\u002F{run_id}\u002Feval_summary.json\n","bash","",[185],{"type":57,"tag":72,"props":186,"children":187},{"__ignoreMap":183},[188,212],{"type":57,"tag":189,"props":190,"children":193},"span",{"class":191,"line":192},"line",1,[194,200,206],{"type":57,"tag":189,"props":195,"children":197},{"style":196},"--shiki-light:#E2931D;--shiki-default:#FFCB6B;--shiki-dark:#FFCB6B",[198],{"type":63,"value":199},"python3",{"type":57,"tag":189,"props":201,"children":203},{"style":202},"--shiki-light:#91B859;--shiki-default:#C3E88D;--shiki-dark:#C3E88D",[204],{"type":63,"value":205}," tools\u002Fagent-eval\u002Fskills\u002Fagent-eval\u002Fscripts\u002Fparse_eval_summary.py",{"type":57,"tag":189,"props":207,"children":209},{"style":208},"--shiki-light:#90A4AE;--shiki-default:#EEFFFF;--shiki-dark:#BABED8",[210],{"type":63,"value":211}," \\\n",{"type":57,"tag":189,"props":213,"children":215},{"class":191,"line":214},2,[216,221],{"type":57,"tag":189,"props":217,"children":218},{"style":202},[219],{"type":63,"value":220},"  --summary-path",{"type":57,"tag":189,"props":222,"children":223},{"style":202},[224],{"type":63,"value":225}," tests\u002Feval\u002Fresults\u002F{run_id}\u002Feval_summary.json\n",{"type":57,"tag":166,"props":227,"children":229},{"id":228},"step-2-perform-the-dialogue-audit-protocol",[230],{"type":63,"value":231},"Step 2: Perform the Dialogue Audit Protocol",{"type":57,"tag":66,"props":233,"children":234},{},[235,237,242],{"type":63,"value":236},"Inspect ",{"type":57,"tag":72,"props":238,"children":240},{"className":239},[],[241],{"type":63,"value":142},{"type":63,"value":243}," for each scenario, analyzing across:",{"type":57,"tag":119,"props":245,"children":246},{},[247,258,268,309],{"type":57,"tag":123,"props":248,"children":249},{},[250,256],{"type":57,"tag":251,"props":252,"children":253},"strong",{},[254],{"type":63,"value":255},"Turn-by-Turn User Intent",{"type":63,"value":257},": What the user requested in Turn 1, Turn 2, etc.",{"type":57,"tag":123,"props":259,"children":260},{},[261,266],{"type":57,"tag":251,"props":262,"children":263},{},[264],{"type":63,"value":265},"Tool Execution & Traceability",{"type":63,"value":267},": Did the agent emit SQL\u002FAPI tool calls on each turn where factual assertions were made, or did it answer from empirical dialogue memory?",{"type":57,"tag":123,"props":269,"children":270},{},[271,276,278],{"type":57,"tag":251,"props":272,"children":273},{},[274],{"type":63,"value":275},"The Memory Reuse vs. Traceability Rubric Clash",{"type":63,"value":277},":\n",{"type":57,"tag":279,"props":280,"children":281},"ul",{},[282],{"type":57,"tag":123,"props":283,"children":284},{},[285,291,293,299,301,307],{"type":57,"tag":286,"props":287,"children":288},"em",{},[289],{"type":63,"value":290},"Diagnostic Pattern",{"type":63,"value":292},": If ",{"type":57,"tag":72,"props":294,"children":296},{"className":295},[],[297],{"type":63,"value":298},"tool_use_quality",{"type":63,"value":300}," drops to ",{"type":57,"tag":72,"props":302,"children":304},{"className":303},[],[305],{"type":63,"value":306},"0.00",{"type":63,"value":308}," on Turn 2 follow-ups, verify if the agent answered correctly from dialogue memory without re-querying the backend. If the rubric expects a tool call on every turn, calibrate the rubric or prompt instructions accordingly.",{"type":57,"tag":123,"props":310,"children":311},{},[312,317],{"type":57,"tag":251,"props":313,"children":314},{},[315],{"type":63,"value":316},"Judge Explanation Grounding",{"type":63,"value":318},": Cross-reference any rubric score $\u003C 1.00$ against the exact model response to verify whether the deduction was a genuine agent failure (Tier 1) or an overly strict judge rubric (Tier 2).",{"type":57,"tag":80,"props":320,"children":321},{},[],{"type":57,"tag":84,"props":323,"children":325},{"id":324},"_3-executive-output-template",[326],{"type":63,"value":327},"3. Executive Output Template",{"type":57,"tag":66,"props":329,"children":330},{},[331],{"type":63,"value":332},"Conclude the audit with a structured summary table:",{"type":57,"tag":334,"props":335,"children":336},"table",{},[337,367],{"type":57,"tag":338,"props":339,"children":340},"thead",{},[341],{"type":57,"tag":342,"props":343,"children":344},"tr",{},[345,352,357,362],{"type":57,"tag":346,"props":347,"children":349},"th",{"align":348},"left",[350],{"type":63,"value":351},"Q#",{"type":57,"tag":346,"props":353,"children":354},{"align":348},[355],{"type":63,"value":356},"Scenario ID",{"type":57,"tag":346,"props":358,"children":359},{"align":348},[360],{"type":63,"value":361},"Metrics Summary",{"type":57,"tag":346,"props":363,"children":364},{"align":348},[365],{"type":63,"value":366},"Root-Cause Finding",{"type":57,"tag":368,"props":369,"children":370},"tbody",{},[371,417],{"type":57,"tag":342,"props":372,"children":373},{},[374,384,393,412],{"type":57,"tag":375,"props":376,"children":377},"td",{"align":348},[378],{"type":57,"tag":72,"props":379,"children":381},{"className":380},[],[382],{"type":63,"value":383},"Q1",{"type":57,"tag":375,"props":385,"children":386},{"align":348},[387],{"type":57,"tag":72,"props":388,"children":390},{"className":389},[],[391],{"type":63,"value":392},"TC-001",{"type":57,"tag":375,"props":394,"children":395},{"align":348},[396,402,404,410],{"type":57,"tag":72,"props":397,"children":399},{"className":398},[],[400],{"type":63,"value":401},"tool_use",{"type":63,"value":403},"=1.0, ",{"type":57,"tag":72,"props":405,"children":407},{"className":406},[],[408],{"type":63,"value":409},"business_logic",{"type":63,"value":411},"=1.0",{"type":57,"tag":375,"props":413,"children":414},{"align":348},[415],{"type":63,"value":416},"✅ PASS (Strict parameter adherence)",{"type":57,"tag":342,"props":418,"children":419},{},[420,429,438,454],{"type":57,"tag":375,"props":421,"children":422},{"align":348},[423],{"type":57,"tag":72,"props":424,"children":426},{"className":425},[],[427],{"type":63,"value":428},"Q2",{"type":57,"tag":375,"props":430,"children":431},{"align":348},[432],{"type":57,"tag":72,"props":433,"children":435},{"className":434},[],[436],{"type":63,"value":437},"TC-002",{"type":57,"tag":375,"props":439,"children":440},{"align":348},[441,446,448,453],{"type":57,"tag":72,"props":442,"children":444},{"className":443},[],[445],{"type":63,"value":401},{"type":63,"value":447},"=0.0, ",{"type":57,"tag":72,"props":449,"children":451},{"className":450},[],[452],{"type":63,"value":409},{"type":63,"value":411},{"type":57,"tag":375,"props":455,"children":456},{"align":348},[457],{"type":63,"value":458},"⚠️ Memory reuse on Turn 2; rubric expected re-query",{"type":57,"tag":460,"props":461,"children":462},"style",{},[463],{"type":63,"value":464},"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":466,"total":499},[467,481,492],{"slug":52,"name":52,"fn":468,"description":469,"org":470,"tags":471,"stars":24,"repoUrl":25,"updatedAt":480},"execute high-performance agent evaluations","Executes high-performance agent evaluations, multi-turn UserSim simulations, and declarative metric grading aligned with google\u002Fagents-cli and the Quality Flywheel. Publishes benchmark artifacts to the GCS Evaluation Registry, executes automated head-to-head delta comparisons (--compare-to), and optimizes system instructions via ADK GEPA (Genetic Evolutionary Prompt Optimization). Use when running agent benchmarks, evaluating ADK\u002FFastAPI agents, diagnosing loss clusters, comparing prompt iterations, running GEPA prompt optimization, or serving evaluation dashboards. Don't use for raw agent code scaffolding (use google-agents-cli-scaffold) or infrastructure deployment (use google-agents-cli-deploy).",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[472,475,478,479],{"name":473,"slug":474,"type":16},"Agents","agents",{"name":476,"slug":477,"type":16},"Benchmarking","benchmarking",{"name":14,"slug":15,"type":16},{"name":9,"slug":8,"type":16},"2026-08-13T05:04:15.77511",{"slug":482,"name":482,"fn":483,"description":484,"org":485,"tags":486,"stars":24,"repoUrl":25,"updatedAt":491},"agent-eval-workflow","design and run agent evaluation workflows","This skill should be used when the user wants to evaluate an AI agent end-to-end: scaffold an evaluation, design metrics that test a real hypothesis, make an agent measurable, audit generated eval config, read evaluation results, or run an improvement (\"hill climbing\") loop. Covers evaluation methodology, metric design, dataset coverage, reading deterministic vs LLM-judged metrics, and the traps that make eval runs silently measure nothing. Use alongside the tool-specific skills (agents-cli-eval, adk-eval-guide) — those cover commands and schemas, this covers the process and judgement.\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[487,488,489,490],{"name":473,"slug":474,"type":16},{"name":476,"slug":477,"type":16},{"name":14,"slug":15,"type":16},{"name":9,"slug":8,"type":16},"2026-08-13T05:04:18.890378",{"slug":4,"name":4,"fn":5,"description":6,"org":493,"tags":494,"stars":24,"repoUrl":25,"updatedAt":26},{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[495,496,497,498],{"name":18,"slug":19,"type":16},{"name":22,"slug":23,"type":16},{"name":14,"slug":15,"type":16},{"name":9,"slug":8,"type":16},3,{"items":501,"total":665},[502,520,536,543,550,557,579,593,604,618,635,652],{"slug":503,"name":503,"fn":504,"description":505,"org":506,"tags":507,"stars":517,"repoUrl":518,"updatedAt":519},"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},[508,511,514],{"name":509,"slug":510,"type":16},"Documentation","documentation",{"name":512,"slug":513,"type":16},"Knowledge Base","knowledge-base",{"name":515,"slug":516,"type":16},"Search","search",8409,"https:\u002F\u002Fgithub.com\u002FGoogleCloudPlatform\u002Fknowledge-catalog","2026-07-12T07:38:52.157375",{"slug":521,"name":522,"fn":523,"description":524,"org":525,"tags":526,"stars":517,"repoUrl":518,"updatedAt":535},"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},[527,530,531,534],{"name":528,"slug":529,"type":16},"Data Analysis","data-analysis",{"name":9,"slug":8,"type":16},{"name":532,"slug":533,"type":16},"Knowledge Management","knowledge-management",{"name":515,"slug":516,"type":16},"2026-07-12T07:38:22.196851",{"slug":52,"name":52,"fn":468,"description":469,"org":537,"tags":538,"stars":24,"repoUrl":25,"updatedAt":480},{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[539,540,541,542],{"name":473,"slug":474,"type":16},{"name":476,"slug":477,"type":16},{"name":14,"slug":15,"type":16},{"name":9,"slug":8,"type":16},{"slug":482,"name":482,"fn":483,"description":484,"org":544,"tags":545,"stars":24,"repoUrl":25,"updatedAt":491},{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[546,547,548,549],{"name":473,"slug":474,"type":16},{"name":476,"slug":477,"type":16},{"name":14,"slug":15,"type":16},{"name":9,"slug":8,"type":16},{"slug":4,"name":4,"fn":5,"description":6,"org":551,"tags":552,"stars":24,"repoUrl":25,"updatedAt":26},{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[553,554,555,556],{"name":18,"slug":19,"type":16},{"name":22,"slug":23,"type":16},{"name":14,"slug":15,"type":16},{"name":9,"slug":8,"type":16},{"slug":558,"name":558,"fn":559,"description":560,"org":561,"tags":562,"stars":576,"repoUrl":577,"updatedAt":578},"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},[563,566,569,572,573],{"name":564,"slug":565,"type":16},"Automation","automation",{"name":567,"slug":568,"type":16},"Engineering","engineering",{"name":570,"slug":571,"type":16},"GitHub","github",{"name":9,"slug":8,"type":16},{"name":574,"slug":575,"type":16},"Pull Requests","pull-requests",2077,"https:\u002F\u002Fgithub.com\u002FGoogleCloudPlatform\u002Fcloud-foundation-fabric","2026-07-31T06:23:36.935005",{"slug":580,"name":580,"fn":581,"description":582,"org":583,"tags":584,"stars":576,"repoUrl":577,"updatedAt":592},"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},[585,586,589],{"name":9,"slug":8,"type":16},{"name":587,"slug":588,"type":16},"Infrastructure as Code","infrastructure-as-code",{"name":590,"slug":591,"type":16},"Terraform","terraform","2026-08-10T04:16:46.817883",{"slug":594,"name":594,"fn":595,"description":596,"org":597,"tags":598,"stars":576,"repoUrl":577,"updatedAt":603},"fast-prerequisites","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},[599,600],{"name":9,"slug":8,"type":16},{"name":601,"slug":602,"type":16},"Operations","operations","2026-08-06T05:36:21.590622",{"slug":605,"name":605,"fn":606,"description":607,"org":608,"tags":609,"stars":615,"repoUrl":616,"updatedAt":617},"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},[610,613,614],{"name":611,"slug":612,"type":16},"CLI","cli",{"name":567,"slug":568,"type":16},{"name":9,"slug":8,"type":16},1178,"https:\u002F\u002Fgithub.com\u002FGoogleCloudPlatform\u002Fvertex-ai-creative-studio","2026-07-12T07:39:08.41406",{"slug":619,"name":619,"fn":620,"description":621,"org":622,"tags":623,"stars":615,"repoUrl":616,"updatedAt":634},"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},[624,627,628,631],{"name":625,"slug":626,"type":16},"API Development","api-development",{"name":9,"slug":8,"type":16},{"name":629,"slug":630,"type":16},"LLM","llm",{"name":632,"slug":633,"type":16},"MCP","mcp","2026-07-12T07:39:10.911302",{"slug":636,"name":636,"fn":637,"description":638,"org":639,"tags":640,"stars":615,"repoUrl":616,"updatedAt":651},"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},[641,644,647,648],{"name":642,"slug":643,"type":16},"Audio","audio",{"name":645,"slug":646,"type":16},"Creative","creative",{"name":9,"slug":8,"type":16},{"name":649,"slug":650,"type":16},"Vertex AI","vertex-ai","2026-07-12T07:39:16.623879",{"slug":653,"name":653,"fn":654,"description":655,"org":656,"tags":657,"stars":615,"repoUrl":616,"updatedAt":664},"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},[658,659,660,663],{"name":645,"slug":646,"type":16},{"name":9,"slug":8,"type":16},{"name":661,"slug":662,"type":16},"Image Generation","image-generation",{"name":649,"slug":650,"type":16},"2026-07-12T07:39:15.372822",84]