[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6GWQZPVZQ40HEK81PARRAM6":3,"skill-same:01M6GWQZPVZQ40HEK81PARRAM6":75},{"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":16,"sourceHtmlUrl":17,"categories":18,"keywords":19,"publishedAt":30,"linkOnly":31,"markdownContent":32,"markdownBody":33,"frontMatter":34,"refName":38,"commitSha":39,"skillDirectory":40,"forksCount":41,"repositoryHtmlUrl":42,"repositoryHomepageUrl":6,"ownerHtmlUrl":43,"ownerBlogUrl":44,"ownerDescription":45,"updatedAt":46,"install":47,"resources":60},"01M6GWQZPVZQ40HEK81PARRAM6","memory-tiering",null,"When saving\u002Fpruning memory, tier facts by relevance.","jiayaoqijia\u002Fcryptoskill","jiayaoqijia","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F1453879?v=4","01M6GMYTX7Q7S504VC6E756ENR","YQ","01M6GMYV1QKJ1CC75MTTGT9PR0",81,"MIT","https:\u002F\u002Fgithub.com\u002Fjiayaoqijia\u002Fcryptoskill\u002Fblob\u002Fmain\u002Fskills\u002Fmcp-servers\u002Fgentech-labs-memory-tiering\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fjiayaoqijia\u002Fcryptoskill\u002Fblob\u002Fmain\u002Fskills\u002Fmcp-servers\u002Fgentech-labs-memory-tiering\u002FSKILL.md",[],[20,21,22,23,24,25,26,27,28,29],"memory","memory tiering","tiering","agentic","context","behavior","facts","agentic-behavior","fleet","when","2026-11-10 10:12:30",false,"---\nname: memory-tiering\ndescription: \"When saving\u002Fpruning memory, tier facts by relevance.\"\nversion: 1.0.0\nauthor: Gentech\nlicense: MIT\ntags: [memory, context, tiering, agentic-behavior, fleet]\n---\n\n# Memory Tiering — Intelligence for Intelligence\n\nJordan's model (Aug 20, 2026): an agent's always-loaded memory (MEMORY.md \u002F USER.md)\nshould NOT be a dump. It should be a **curated index** of the facts we touch every\nsession, with everything else parked in the vault brain where it's one search away.\n\n## When to Use\n- Before saving a NEW durable fact to persistent memory.\n- When pruning \u002F consolidating MEMORY.md or USER.md (including via the fleet dietician).\n- When you notice a memory file approaching the 85% threshold.\n- When a fact becomes RESOLVED and you decide whether to keep it loaded.\n\nTwo tiers:\n- **Always-loaded** (MEMORY.md \u002F USER.md) — facts referenced EVERY session. Identity,\n  rails, wallet, rules, standing strategy. Lean, high-signal.\n- **Vault brain** (retrievable) — facts referenced less often, or RESOLVED. Not lost,\n  just parked. Pull back via session_search \u002F search_files when the topic resurfaces.\n\n## Why (the intelligence part)\n- Every session reads the memory file. Bloat = slower, more tokens, worse focus.\n- **More memory = worse performance** (proven Jul 28, 2026: elaborate memory stack\n  deleted → agent worked dramatically better).\n- Trimming from the BACK (oldest-first) is dumb — an old high-frequency fact gets\n  archived while a recent resolved worry stays loaded. Tier by **relevance**, not age.\n\n## Tiering rules\n\n### ALWAYS-LOADED (keep in MEMORY.md\u002FUSER.md)\n- **Identity**: who the agent is, personality, speech, the boss.\n- **Core rails\u002Fkeys**: payment rails, wallets, canonical repos, the strategy that\n  shapes every decision.\n- **Standing rules\u002Fred lines**: \"never fake receipts,\" \"spend less,\" \"flag blockers\n  immediately,\" \"build first talk later.\"\n- **Recurring topics** (per Jordan Aug 20): agentic treasury, x402, AWS, APIs.\n- **Active commitments with near deadlines.**\n\n### ARCHIVES TO VAULT BRAIN (low-frequency \u002F resolved)\n- **Resolved worries**: problem solved, deadline passed, task done. Compress or archive.\n- **One-off events**: a single completed hackathon\u002Fhire\u002Fdate.\n- **Task progress \u002F session logs**: go to the brain (Green Room \u002F Mess Hall \u002F handoffs),\n  NOT persistent memory.\n- **Any fact not touched in many sessions** and not needed for the standing model.\n\n## How to apply (standing agentic behavior — every save)\n1. Before adding memory, ask: \"needed EVERY session, or only when this topic comes\n   up?\" Only-on-topic → save to the Vault instead.\n2. When a fact is RESOLVED: compress to a one-liner or archive — don't keep the\n   full narrative loaded.\n3. At ~85% threshold: consolidate in ONE atomic batch (remove stale + add new),\n   never cram.\n4. Fleet dietician (memory-dietician.py) tiers by relevance, not back-of-file.\n5. When a topic resurfaces, pull the archived entry back from the vault and re-load\n   it if it's now a standing fact.\n\n## Pitfalls\n- Don't archive on a whim — if referenced across sessions, keep it.\n- Don't keep resolved worries loaded \"just in case\" — archive them; the vault has room.\n- The vault brain has room — that's its purpose. Overflow belongs there.\n- This is a STANDING behavior, applied on every save, not a one-time cleanup.\n","# Memory Tiering — Intelligence for Intelligence\n\nJordan's model (Aug 20, 2026): an agent's always-loaded memory (MEMORY.md \u002F USER.md)\nshould NOT be a dump. It should be a **curated index** of the facts we touch every\nsession, with everything else parked in the vault brain where it's one search away.\n\n## When to Use\n- Before saving a NEW durable fact to persistent memory.\n- When pruning \u002F consolidating MEMORY.md or USER.md (including via the fleet dietician).\n- When you notice a memory file approaching the 85% threshold.\n- When a fact becomes RESOLVED and you decide whether to keep it loaded.\n\nTwo tiers:\n- **Always-loaded** (MEMORY.md \u002F USER.md) — facts referenced EVERY session. Identity,\n  rails, wallet, rules, standing strategy. Lean, high-signal.\n- **Vault brain** (retrievable) — facts referenced less often, or RESOLVED. Not lost,\n  just parked. Pull back via session_search \u002F search_files when the topic resurfaces.\n\n## Why (the intelligence part)\n- Every session reads the memory file. Bloat = slower, more tokens, worse focus.\n- **More memory = worse performance** (proven Jul 28, 2026: elaborate memory stack\n  deleted → agent worked dramatically better).\n- Trimming from the BACK (oldest-first) is dumb — an old high-frequency fact gets\n  archived while a recent resolved worry stays loaded. Tier by **relevance**, not age.\n\n## Tiering rules\n\n### ALWAYS-LOADED (keep in MEMORY.md\u002FUSER.md)\n- **Identity**: who the agent is, personality, speech, the boss.\n- **Core rails\u002Fkeys**: payment rails, wallets, canonical repos, the strategy that\n  shapes every decision.\n- **Standing rules\u002Fred lines**: \"never fake receipts,\" \"spend less,\" \"flag blockers\n  immediately,\" \"build first talk later.\"\n- **Recurring topics** (per Jordan Aug 20): agentic treasury, x402, AWS, APIs.\n- **Active commitments with near deadlines.**\n\n### ARCHIVES TO VAULT BRAIN (low-frequency \u002F resolved)\n- **Resolved worries**: problem solved, deadline passed, task done. Compress or archive.\n- **One-off events**: a single completed hackathon\u002Fhire\u002Fdate.\n- **Task progress \u002F session logs**: go to the brain (Green Room \u002F Mess Hall \u002F handoffs),\n  NOT persistent memory.\n- **Any fact not touched in many sessions** and not needed for the standing model.\n\n## How to apply (standing agentic behavior — every save)\n1. Before adding memory, ask: \"needed EVERY session, or only when this topic comes\n   up?\" Only-on-topic → save to the Vault instead.\n2. When a fact is RESOLVED: compress to a one-liner or archive — don't keep the\n   full narrative loaded.\n3. At ~85% threshold: consolidate in ONE atomic batch (remove stale + add new),\n   never cram.\n4. Fleet dietician (memory-dietician.py) tiers by relevance, not back-of-file.\n5. When a topic resurfaces, pull the archived entry back from the vault and re-load\n   it if it's now a standing fact.\n\n## Pitfalls\n- Don't archive on a whim — if referenced across sessions, keep it.\n- Don't keep resolved worries loaded \"just in case\" — archive them; the vault has room.\n- The vault brain has room — that's its purpose. Overflow belongs there.\n- This is a STANDING behavior, applied on every save, not a one-time cleanup.\n",{"name":5,"tags":35,"author":36,"license":15,"version":37,"description":7},[20,24,22,27,28],"Gentech","1.0.0","main","3df1d0d67b30ff15236561f8c0f9f06d33f42dfd","skills\u002Fmcp-servers\u002Fgentech-labs-memory-tiering",17,"https:\u002F\u002Fgithub.com\u002Fjiayaoqijia\u002Fcryptoskill","https:\u002F\u002Fgithub.com\u002Fjiayaoqijia","https:\u002F\u002Feth2030.com","Hardcore infra. Building @alt-research and applied cryptography.","2026-11-12 02:28:42",{"source":8,"skillName":5,"commands":48,"prompt":58,"zipAvailable":59},[49,52,55],{"tool":50,"command":51},"npx","npx skills add jiayaoqijia\u002Fcryptoskill --skill memory-tiering",{"tool":53,"command":54},"bunx","bunx skills add jiayaoqijia\u002Fcryptoskill --skill memory-tiering",{"tool":56,"command":57},"pnpm","pnpm dlx skills add jiayaoqijia\u002Fcryptoskill --skill memory-tiering","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fjiayaoqijia\u002Fcryptoskill\u002Fblob\u002Fmain\u002Fskills\u002Fmcp-servers\u002Fgentech-labs-memory-tiering\u002FSKILL.md",true,[61,65,71],{"path":62,"resourceType":62,"resourceTypeDesc":63,"sizeBytes":64,"sourceHtmlUrl":16},"LICENSE","许可证",1068,{"path":66,"resourceType":67,"resourceTypeDesc":68,"sizeBytes":69,"sourceHtmlUrl":70},"SOURCE.md","REFERENCE","参考文档",366,"https:\u002F\u002Fgithub.com\u002Fjiayaoqijia\u002Fcryptoskill\u002Fblob\u002Fmain\u002Fskills\u002Fmcp-servers\u002Fgentech-labs-memory-tiering\u002FSOURCE.md",{"path":72,"resourceType":67,"resourceTypeDesc":68,"sizeBytes":73,"sourceHtmlUrl":74},"TRUST.auto.yaml",2217,"https:\u002F\u002Fgithub.com\u002Fjiayaoqijia\u002Fcryptoskill\u002Fblob\u002Fmain\u002Fskills\u002Fmcp-servers\u002Fgentech-labs-memory-tiering\u002FTRUST.auto.yaml",[76,104,127,145,158,168],{"skillNo":77,"name":78,"nameZh":6,"description":79,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":80,"sourceHtmlUrl":81,"categories":82,"keywords":83,"publishedAt":30},"01M712C436EQYE43342Q4E7GG1","recipe-performance","Diagnose a specific MetaMask user flow's real-app performance and Sentry critical-user-flow telemetry with deterministic mm-harness recipes. Use for CUF boundaries, cold\u002Fwarm\u002Fbackground-resume behavior, latency regressions, lifecycle tags, duplicate or blocking network requests, long tasks, layout\u002Fpaint cost, screen redraws, React rerenders, or browser trace analysis where tests alone cannot prove the real app.","https:\u002F\u002Fgithub.com\u002Fjiayaoqijia\u002Fcryptoskill\u002Fblob\u002Fmain\u002Fskills\u002Fwallets\u002Fmetamask-official-recipe-performance\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fjiayaoqijia\u002Fcryptoskill\u002Fblob\u002Fmain\u002Fskills\u002Fwallets\u002Fmetamask-official-recipe-performance\u002FSKILL.md",[],[84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103],"performance","recipe performance","recipe","mm","sentry","harness","app","deterministic","recipes","flow","cold","critical","analysis","tasks","react","browser","layout","screen","network","metamask",{"skillNo":105,"name":106,"nameZh":6,"description":107,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":108,"licenseFileUrl":109,"sourceHtmlUrl":110,"categories":111,"keywords":112,"publishedAt":30},"01M70XT2F8A5CQZ81JBP98KSW3","mexc-spot-websocket","MEXC Spot WebSocket workflows for public streams, protobuf handling, depth maintenance, listen keys, and private streams without exposing API keys.","Apache-2.0","https:\u002F\u002Fgithub.com\u002Fjiayaoqijia\u002Fcryptoskill\u002Fblob\u002Fmain\u002Fskills\u002Fexchanges\u002Fmncrftfrcnm-mexc-spot-websocket\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fjiayaoqijia\u002Fcryptoskill\u002Fblob\u002Fmain\u002Fskills\u002Fexchanges\u002Fmncrftfrcnm-mexc-spot-websocket\u002FSKILL.md",[],[113,114,115,116,117,118,119,120,121,122,123,124,125,126],"mexc spot websocket","spot","websocket","api","workflows","public","listen","streams","depth","keys","for","api keys","maintenance","protobuf",{"skillNo":128,"name":129,"nameZh":6,"description":130,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":108,"licenseFileUrl":131,"sourceHtmlUrl":132,"categories":133,"keywords":134,"publishedAt":30},"01M70XT0MBXN3WZY1BKN0CPNKR","mexc-spot-rest","MEXC Spot REST workflows for public data, signed account requests, orders, wallet\u002Fadmin actions, and secure signing without exposing API keys.","https:\u002F\u002Fgithub.com\u002Fjiayaoqijia\u002Fcryptoskill\u002Fblob\u002Fmain\u002Fskills\u002Fexchanges\u002Fmncrftfrcnm-mexc-spot-rest\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fjiayaoqijia\u002Fcryptoskill\u002Fblob\u002Fmain\u002Fskills\u002Fexchanges\u002Fmncrftfrcnm-mexc-spot-rest\u002FSKILL.md",[],[135,114,136,137,138,139,140,124,123,122,141,142,143,118,144,117,116],"mexc spot rest","REST","admin","signing","Wallet","requests","actions","secure","data","orders",{"skillNo":146,"name":147,"nameZh":6,"description":148,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":108,"licenseFileUrl":149,"sourceHtmlUrl":150,"categories":151,"keywords":152,"publishedAt":30},"01M70XSYSVZ2N0N91XCAR2CAK0","mexc-futures-websocket","MEXC Futures WebSocket workflows for public streams, private account\u002Forder\u002Fposition streams, ping\u002Fpong, subscriptions, and depth maintenance.","https:\u002F\u002Fgithub.com\u002Fjiayaoqijia\u002Fcryptoskill\u002Fblob\u002Fmain\u002Fskills\u002Fexchanges\u002Fmncrftfrcnm-mexc-futures-websocket\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fjiayaoqijia\u002Fcryptoskill\u002Fblob\u002Fmain\u002Fskills\u002Fexchanges\u002Fmncrftfrcnm-mexc-futures-websocket\u002FSKILL.md",[],[153,154,115,155,156,125,123,157,121,120,118,117],"mexc futures websocket","futures","position","ping","subscriptions",{"skillNo":159,"name":160,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":161,"sourceHtmlUrl":162,"categories":163,"keywords":164,"publishedAt":30},"01M6H0ZWHVE5W04WG2DKQE2SQC","0xklen-version-and-refresh-sanctions-lists","https:\u002F\u002Fgithub.com\u002Fjiayaoqijia\u002Fcryptoskill\u002Fblob\u002Fmain\u002Fskills\u002Fpayments\u002F0xklen-version-and-refresh-sanctions-lists\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fjiayaoqijia\u002Fcryptoskill\u002Fblob\u002Fmain\u002Fskills\u002Fpayments\u002F0xklen-version-and-refresh-sanctions-lists\u002FSKILL.md",[],[165,166,167],"refresh","0xklen version and refresh sanctions lists","version",{"skillNo":169,"name":170,"nameZh":6,"description":171,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":172,"sourceHtmlUrl":173,"categories":174,"keywords":175,"publishedAt":30},"01M6H0Z28Q3R8J0S799GJQWXM3","verify-tests-cover-changed-lines","Use when a PR adds code that the existing suite may not exercise. Measures patch coverage — the share of newly changed lines hit by tests — before approving.","https:\u002F\u002Fgithub.com\u002Fjiayaoqijia\u002Fcryptoskill\u002Fblob\u002Fmain\u002Fskills\u002Fpayments\u002F0xklen-verify-tests-cover-changed-lines\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fjiayaoqijia\u002Fcryptoskill\u002Fblob\u002Fmain\u002Fskills\u002Fpayments\u002F0xklen-verify-tests-cover-changed-lines\u002FSKILL.md",[],[176,177,178,179,180,181,182,183,184,185,186],"tests","cover","verify tests cover changed lines","verify","coverage","pr","exercise","use","patch","code","share"]