[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7BWCGPJR7K9W23P01DVX5NF":3,"skill-same:01M7BWCGPJR7K9W23P01DVX5NF":49},{"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":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":27,"commitSha":28,"skillDirectory":29,"forksCount":30,"repositoryHtmlUrl":31,"repositoryHomepageUrl":6,"ownerHtmlUrl":32,"ownerBlogUrl":6,"ownerDescription":33,"updatedAt":34,"install":35,"resources":48},"01M7BWCGPJR7K9W23P01DVX5NF","clickhouse-managed-postgres-rca",null,"MUST USE when investigating performance issues on a ClickHouse-managed Postgres instance. Provides an evidence-based RCA workflow that scrapes the Prometheus endpoint for system signal, pulls per-digest evidence from the Slow Query Patterns API, and recommends (does not apply) a fix.","om-scogo\u002Fskillsh-scraper","om-scogo","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F160466840?v=4","01M79AWB6DFVCQ7RK70CS34B96","Om Achrekar","01M79AWB6ZM9SKR8KDH1WSSM22",0,"Apache-2.0","https:\u002F\u002Fgithub.com\u002Fom-scogo\u002Fskillsh-scraper\u002Fblob\u002Fmain\u002Fdata\u002Fskills\u002Fclickhouse-agent-skills\u002Fclickhouse-managed-postgres-rca\u002Fskills\u002Fclickhouse-managed-postgres-rca\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: clickhouse-managed-postgres-rca\ndescription: MUST USE when investigating performance issues on a ClickHouse-managed Postgres instance. Provides an evidence-based RCA workflow that scrapes the Prometheus endpoint for system signal, pulls per-digest evidence from the Slow Query Patterns API, and recommends (does not apply) a fix.\nlicense: Apache-2.0\nmetadata:\n  author: ClickHouse Inc\n  version: \"0.1.0\"\n---\n\n# ClickHouse Managed Postgres RCA\n\n## When to use\n\nTrigger whenever a user reports slowness, high CPU, low\nthroughput, cache thrash, or any unexplained pain on a\nClickHouse-managed Postgres instance.\n\n## What you have access to\n\nTwo APIs on `https:\u002F\u002Fapi.clickhouse.cloud` (HTTP Basic auth\nusing a ClickHouse Cloud API key\u002Fsecret pair):\n\n- **Prometheus metrics** — operation `postgresInstancePrometheusGet`\n  under the Prometheus tag. Returns Prometheus exposition format.\n  System and workload metrics for one Postgres service.\n- **Slow Query Patterns** — operation `slowQueryPatternsGetList`\n  under the Postgres tag. Returns per-digest latency, IO, and\n  call statistics for normalized query patterns. **Beta.**\n\nBoth endpoints require an `organizationId` and a `serviceId` as\npath parameters. The user must supply both, plus the API\nkey\u002Fsecret pair.\n\n## What you do NOT have\n\n- Query plans \u002F EXPLAIN output.\n- Per-table scan-type counters (`seq_scan` \u002F `idx_scan`).\n- Autovacuum or last-ANALYZE timestamps.\n\nReason from IO and timing signals, not from a plan tree.\n\n## Workflow\n\nSix steps, in order. Do not skip ahead.\n\nSteps 2 and 3 only share auth — no data dependency between\nthem. Run them in parallel (background curls, `&` + `wait`) to\ncut wall time from sequential ~2s to ~1s.\n\n### 1. Discover the live API shape\n\nThese endpoints are Beta — paths, params, and JSON field names\ncan shift. Follow `rules\u002Fopenapi-discovery.md` to:\n\n1. Fetch the OpenAPI spec from `https:\u002F\u002Fapi.clickhouse.cloud\u002Fv1`.\n2. Locate the two operations by `operationId`:\n   - `postgresInstancePrometheusGet` (Prometheus tag)\n   - `slowQueryPatternsGetList` (Postgres tag)\n3. Resolve their path templates, required query parameters,\n   and (for the slow-query endpoint) the response schema.\n4. Build a session-scoped role map from the schema property\n   descriptions: `{ semantic role → actual field name }`.\n\nUse the resolved names in every subsequent request and citation.\nNever hardcode field names from memory.\n\n### 2. Scrape Prom once for system gauges\n\nFollow `rules\u002Fprometheus-scrape.md`. **One scrape, no wait.**\nYou're after gauges (current values) that don't need a delta:\n`CacheHitRatio`, `ActiveConnections`, `MemoryUsedPercent`,\n`FilesystemUsedPercent`.\n\nA `CacheHitRatio` well below ~95% on a workload that should\nfit in cache is a real signal on its own. Climbing\n`ActiveConnections` toward the pool ceiling is a real signal\non its own. These don't need rate-of-change.\n\nA second scrape for counter deltas is **opt-in**, used only\nwhen Step 4 triage points at write-congestion (where deadlock\nand rollback *rates* matter and the Slow Query Patterns API\ncan't substitute). For the read-path case (the most common\nRCA shape) the single scrape is enough.\n\n### 3. Pull top slow query patterns\n\nRequest the slow query patterns. Follow\n`rules\u002Fslow-query-patterns-fields.md` for the fields that\nmatter and how to read them. This is the primary diagnostic —\nit returns per-pattern accumulated totals (call count, runtime,\nblocks, rows) over the window you request, which is the\n\"rate-of-change\" data you'd otherwise derive from two Prom\nscrapes — but per query and without waiting.\n\nIf no patterns return a meaningful `totalDurationUs`, the\nreport may be overstated or the issue isn't query-shaped.\nStop and tell the user what you looked at.\n\n### 4. Triage: pick the right heuristic\n\nFollow `rules\u002Ftriage.md`. Match the combined Prom + slow-query\nsignal to one of the heuristic shapes. Each shape points to a\nspecific heuristic file:\n\n- `rules\u002Fheuristic-full-scan.md` — read-path full scan.\n- `rules\u002Fheuristic-hot-loop.md` — N+1 \u002F hot loop from the app.\n- `rules\u002Fheuristic-write-congestion.md` — deadlocks, slow\n  writes, high rollback rate.\n\nIf the signal does not match any shape cleanly, do not invent\na hypothesis. Surface the top patterns and ask the user which\nworkload they recognize. New heuristics are welcome as PRs.\n\n### 5. Reason, then recommend\n\nUse the format in `rules\u002Foutput-template.md`. Always include:\nsymptom, evidence, hypothesis (noting any alternative cause\nyou cannot rule out from this surface alone), short-term fix,\nand long-term follow-ups.\n\n### 6. Do not apply the fix\n\nFollow `rules\u002Frecommend-only.md`. Never run DDL. Never call\n`pg_cancel_backend` or `pg_terminate_backend`. Write the\nrecommendation, explain why, and let the human apply it.\n\n## Full Compiled Document\n\nFor the complete guide with every rule expanded in a single\ncontext load: `AGENTS.md`.\n","# ClickHouse Managed Postgres RCA\n\n## When to use\n\nTrigger whenever a user reports slowness, high CPU, low\nthroughput, cache thrash, or any unexplained pain on a\nClickHouse-managed Postgres instance.\n\n## What you have access to\n\nTwo APIs on `https:\u002F\u002Fapi.clickhouse.cloud` (HTTP Basic auth\nusing a ClickHouse Cloud API key\u002Fsecret pair):\n\n- **Prometheus metrics** — operation `postgresInstancePrometheusGet`\n  under the Prometheus tag. Returns Prometheus exposition format.\n  System and workload metrics for one Postgres service.\n- **Slow Query Patterns** — operation `slowQueryPatternsGetList`\n  under the Postgres tag. Returns per-digest latency, IO, and\n  call statistics for normalized query patterns. **Beta.**\n\nBoth endpoints require an `organizationId` and a `serviceId` as\npath parameters. The user must supply both, plus the API\nkey\u002Fsecret pair.\n\n## What you do NOT have\n\n- Query plans \u002F EXPLAIN output.\n- Per-table scan-type counters (`seq_scan` \u002F `idx_scan`).\n- Autovacuum or last-ANALYZE timestamps.\n\nReason from IO and timing signals, not from a plan tree.\n\n## Workflow\n\nSix steps, in order. Do not skip ahead.\n\nSteps 2 and 3 only share auth — no data dependency between\nthem. Run them in parallel (background curls, `&` + `wait`) to\ncut wall time from sequential ~2s to ~1s.\n\n### 1. Discover the live API shape\n\nThese endpoints are Beta — paths, params, and JSON field names\ncan shift. Follow `rules\u002Fopenapi-discovery.md` to:\n\n1. Fetch the OpenAPI spec from `https:\u002F\u002Fapi.clickhouse.cloud\u002Fv1`.\n2. Locate the two operations by `operationId`:\n   - `postgresInstancePrometheusGet` (Prometheus tag)\n   - `slowQueryPatternsGetList` (Postgres tag)\n3. Resolve their path templates, required query parameters,\n   and (for the slow-query endpoint) the response schema.\n4. Build a session-scoped role map from the schema property\n   descriptions: `{ semantic role → actual field name }`.\n\nUse the resolved names in every subsequent request and citation.\nNever hardcode field names from memory.\n\n### 2. Scrape Prom once for system gauges\n\nFollow `rules\u002Fprometheus-scrape.md`. **One scrape, no wait.**\nYou're after gauges (current values) that don't need a delta:\n`CacheHitRatio`, `ActiveConnections`, `MemoryUsedPercent`,\n`FilesystemUsedPercent`.\n\nA `CacheHitRatio` well below ~95% on a workload that should\nfit in cache is a real signal on its own. Climbing\n`ActiveConnections` toward the pool ceiling is a real signal\non its own. These don't need rate-of-change.\n\nA second scrape for counter deltas is **opt-in**, used only\nwhen Step 4 triage points at write-congestion (where deadlock\nand rollback *rates* matter and the Slow Query Patterns API\ncan't substitute). For the read-path case (the most common\nRCA shape) the single scrape is enough.\n\n### 3. Pull top slow query patterns\n\nRequest the slow query patterns. Follow\n`rules\u002Fslow-query-patterns-fields.md` for the fields that\nmatter and how to read them. This is the primary diagnostic —\nit returns per-pattern accumulated totals (call count, runtime,\nblocks, rows) over the window you request, which is the\n\"rate-of-change\" data you'd otherwise derive from two Prom\nscrapes — but per query and without waiting.\n\nIf no patterns return a meaningful `totalDurationUs`, the\nreport may be overstated or the issue isn't query-shaped.\nStop and tell the user what you looked at.\n\n### 4. Triage: pick the right heuristic\n\nFollow `rules\u002Ftriage.md`. Match the combined Prom + slow-query\nsignal to one of the heuristic shapes. Each shape points to a\nspecific heuristic file:\n\n- `rules\u002Fheuristic-full-scan.md` — read-path full scan.\n- `rules\u002Fheuristic-hot-loop.md` — N+1 \u002F hot loop from the app.\n- `rules\u002Fheuristic-write-congestion.md` — deadlocks, slow\n  writes, high rollback rate.\n\nIf the signal does not match any shape cleanly, do not invent\na hypothesis. Surface the top patterns and ask the user which\nworkload they recognize. New heuristics are welcome as PRs.\n\n### 5. Reason, then recommend\n\nUse the format in `rules\u002Foutput-template.md`. Always include:\nsymptom, evidence, hypothesis (noting any alternative cause\nyou cannot rule out from this surface alone), short-term fix,\nand long-term follow-ups.\n\n### 6. Do not apply the fix\n\nFollow `rules\u002Frecommend-only.md`. Never run DDL. Never call\n`pg_cancel_backend` or `pg_terminate_backend`. Write the\nrecommendation, explain why, and let the human apply it.\n\n## Full Compiled Document\n\nFor the complete guide with every rule expanded in a single\ncontext load: `AGENTS.md`.\n",{"name":5,"license":15,"metadata":24,"description":7},{"author":25,"version":26},"ClickHouse Inc","0.1.0","main","8859218955d9504cd18b59e7892100718039d3b4","data\u002Fskills\u002Fclickhouse-agent-skills\u002Fclickhouse-managed-postgres-rca\u002Fskills\u002Fclickhouse-managed-postgres-rca",1,"https:\u002F\u002Fgithub.com\u002Fom-scogo\u002Fskillsh-scraper","https:\u002F\u002Fgithub.com\u002Fom-scogo","SDE-1 @ Scogo ","2026-07-18 12:22:46",{"source":8,"skillName":5,"commands":36,"prompt":46,"zipAvailable":47},[37,40,43],{"tool":38,"command":39},"npx","npx skills add om-scogo\u002Fskillsh-scraper --skill clickhouse-managed-postgres-rca",{"tool":41,"command":42},"bunx","bunx skills add om-scogo\u002Fskillsh-scraper --skill clickhouse-managed-postgres-rca",{"tool":44,"command":45},"pnpm","pnpm dlx skills add om-scogo\u002Fskillsh-scraper --skill clickhouse-managed-postgres-rca","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fom-scogo\u002Fskillsh-scraper\u002Fblob\u002Fmain\u002Fdata\u002Fskills\u002Fclickhouse-agent-skills\u002Fclickhouse-managed-postgres-rca\u002Fskills\u002Fclickhouse-managed-postgres-rca\u002FSKILL.md",true,[],[50,58,63,70,75,82],{"skillNo":51,"name":52,"nameZh":6,"description":53,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":54,"licenseFileUrl":6,"sourceHtmlUrl":55,"categories":56,"keywords":57,"publishedAt":19},"01M7D6MC9FAAEV4CMNYX8W7JTG","arxiv-search","Search arXiv physics, math, and computer science preprints using natural language queries. Powered by Valyu semantic search.","MIT","https:\u002F\u002Fgithub.com\u002Fom-scogo\u002Fskillsh-scraper\u002Fblob\u002Fmain\u002Fdata\u002Fskills\u002Fyorkeccak-scientific-skills\u002Farxiv-search\u002Fskills\u002Farxiv-search\u002FSKILL.md",[],[],{"skillNo":59,"name":52,"nameZh":6,"description":53,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":54,"licenseFileUrl":6,"sourceHtmlUrl":60,"categories":61,"keywords":62,"publishedAt":19},"01M7D6M85CKKN87TA98VCQV4H7","https:\u002F\u002Fgithub.com\u002Fom-scogo\u002Fskillsh-scraper\u002Fblob\u002Fmain\u002Fdata\u002Fskills\u002Fyorkeccak-scientific-skills\u002Farxiv-search\u002FSkill.md",[],[],{"skillNo":64,"name":65,"nameZh":6,"description":66,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":54,"licenseFileUrl":6,"sourceHtmlUrl":67,"categories":68,"keywords":69,"publishedAt":19},"01M7D4ZTCS4DR3CE4JY06X84T2","working-with-legacy-code","Safely change and test untested codebases using Feathers' \"Working Effectively with Legacy Code\". Use when the user mentions \"legacy code\", \"no tests\", \"untested codebase\", \"how do I test this\", \"seams\", \"characterization tests\", \"golden master\", \"sprout method\", \"afraid to change this code\", \"monster method\", \"dependency breaking\", or \"inherited a messy codebase\". Also trigger when changing code without tests safely, getting a class under test when constructors, statics, or singletons block it, adding features to tangled modules, or planning incremental test coverage for an old codebase. Covers the legacy-code change algorithm, seams, characterization tests, sprout\u002Fwrap, and dependency-breaking techniques. For refactoring code that already has tests, see refactoring-patterns. For day-to-day code quality, see clean-code.","https:\u002F\u002Fgithub.com\u002Fom-scogo\u002Fskillsh-scraper\u002Fblob\u002Fmain\u002Fdata\u002Fskills\u002Fwondelai-skills\u002Fworking-with-legacy-code\u002Fworking-with-legacy-code\u002FSKILL.md",[],[],{"skillNo":71,"name":65,"nameZh":6,"description":66,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":54,"licenseFileUrl":6,"sourceHtmlUrl":72,"categories":73,"keywords":74,"publishedAt":19},"01M7D4ZHZKJD2RSNGAF63V14P4","https:\u002F\u002Fgithub.com\u002Fom-scogo\u002Fskillsh-scraper\u002Fblob\u002Fmain\u002Fdata\u002Fskills\u002Fwondelai-skills\u002Fworking-with-legacy-code\u002FSkill.md",[],[],{"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":54,"licenseFileUrl":6,"sourceHtmlUrl":79,"categories":80,"keywords":81,"publishedAt":19},"01M7D4ZB4CJBFW5JE67CPH2QRV","web-typography","Select, pair, and implement typefaces for web projects. Use when the user mentions \"font pairing\", \"which typeface\", \"line height\", \"responsive typography\", \"web font loading\", \"type hierarchy\", \"variable fonts\", \"FOUT\u002FFOIT\", \"typographic scale\", or \"the text is hard to read\". Also trigger when choosing between system fonts and web fonts, optimizing font-loading performance, or designing readable long-form content. Covers readability evaluation, CSS implementation, and performance optimization. For overall UI design systems, see refactoring-ui. For dramatic typographic experiences, see top-design.","https:\u002F\u002Fgithub.com\u002Fom-scogo\u002Fskillsh-scraper\u002Fblob\u002Fmain\u002Fdata\u002Fskills\u002Fwondelai-skills\u002Fweb-typography\u002Fweb-typography\u002FSKILL.md",[],[],{"skillNo":83,"name":77,"nameZh":6,"description":78,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":54,"licenseFileUrl":6,"sourceHtmlUrl":84,"categories":85,"keywords":86,"publishedAt":19},"01M7D4Z4MVYVY3NMJ18BNV6WNA","https:\u002F\u002Fgithub.com\u002Fom-scogo\u002Fskillsh-scraper\u002Fblob\u002Fmain\u002Fdata\u002Fskills\u002Fwondelai-skills\u002Fweb-typography\u002FSkill.md",[],[]]