[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6NV6P9ZDCEY7QHKYBAP8P40":3,"skill-same:01M6NV6P9ZDCEY7QHKYBAP8P40":50},{"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":23,"linkOnly":24,"markdownContent":25,"markdownBody":26,"frontMatter":27,"refName":28,"commitSha":29,"skillDirectory":30,"forksCount":14,"repositoryHtmlUrl":31,"repositoryHomepageUrl":6,"ownerHtmlUrl":32,"ownerBlogUrl":33,"ownerDescription":34,"updatedAt":35,"install":36,"resources":49},"01M6NV6P9ZDCEY7QHKYBAP8P40","llm-tuning-patterns",null,"LLM Tuning Patterns","rubicanjr\u002FFinCognis","rubicanjr","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F234912519?v=4","01M6NT156NND8J474G34SB3923","Rubi","01M6NT15B70RWXE1N3YV0BZY8A",0,"MIT","https:\u002F\u002Fgithub.com\u002Frubicanjr\u002FFinCognis\u002Fblob\u002Fmain\u002Fskills\u002Fllm-tuning-patterns\u002FSKILL.md",[],[19,20,21,22],"tuning","patterns","llm tuning patterns","false","2026-11-10 10:12:30",false,"---\nname: llm-tuning-patterns\ndescription: LLM Tuning Patterns\nuser-invocable: false\n---\n\n# LLM Tuning Patterns\n\nEvidence-based patterns for configuring LLM parameters, based on APOLLO and Godel-Prover research.\n\n## Pattern\n\nDifferent tasks require different LLM configurations. Use these evidence-based settings.\n\n## Theorem Proving \u002F Formal Reasoning\n\nBased on APOLLO parity analysis:\n\n| Parameter | Value | Rationale |\n|-----------|-------|-----------|\n| max_tokens | 4096 | Proofs need space for chain-of-thought |\n| temperature | 0.6 | Higher creativity for tactic exploration |\n| top_p | 0.95 | Allow diverse proof paths |\n\n### Proof Plan Prompt\n\nAlways request a proof plan before tactics:\n\n```\nGiven the theorem to prove:\n[theorem statement]\n\nFirst, write a high-level proof plan explaining your approach.\nThen, suggest Lean 4 tactics to implement each step.\n```\n\nThe proof plan (chain-of-thought) significantly improves tactic quality.\n\n### Parallel Sampling\n\nFor hard proofs, use parallel sampling:\n- Generate N=8-32 candidate proof attempts\n- Use best-of-N selection\n- Each sample at temperature 0.6-0.8\n\n## Code Generation\n\n| Parameter | Value | Rationale |\n|-----------|-------|-----------|\n| max_tokens | 2048 | Sufficient for most functions |\n| temperature | 0.2-0.4 | Prefer deterministic output |\n\n## Creative \u002F Exploration Tasks\n\n| Parameter | Value | Rationale |\n|-----------|-------|-----------|\n| max_tokens | 4096 | Space for exploration |\n| temperature | 0.8-1.0 | Maximum creativity |\n\n## Anti-Patterns\n\n- **Too low tokens for proofs**: 512 tokens truncates chain-of-thought\n- **Too low temperature for proofs**: 0.2 misses creative tactic paths\n- **No proof plan**: Jumping to tactics without planning reduces success rate\n\n## Source Sessions\n\n- This session: APOLLO parity - increased max_tokens 512->4096, temp 0.2->0.6\n- This session: Added proof plan prompt for chain-of-thought before tactics\n","# LLM Tuning Patterns\n\nEvidence-based patterns for configuring LLM parameters, based on APOLLO and Godel-Prover research.\n\n## Pattern\n\nDifferent tasks require different LLM configurations. Use these evidence-based settings.\n\n## Theorem Proving \u002F Formal Reasoning\n\nBased on APOLLO parity analysis:\n\n| Parameter | Value | Rationale |\n|-----------|-------|-----------|\n| max_tokens | 4096 | Proofs need space for chain-of-thought |\n| temperature | 0.6 | Higher creativity for tactic exploration |\n| top_p | 0.95 | Allow diverse proof paths |\n\n### Proof Plan Prompt\n\nAlways request a proof plan before tactics:\n\n```\nGiven the theorem to prove:\n[theorem statement]\n\nFirst, write a high-level proof plan explaining your approach.\nThen, suggest Lean 4 tactics to implement each step.\n```\n\nThe proof plan (chain-of-thought) significantly improves tactic quality.\n\n### Parallel Sampling\n\nFor hard proofs, use parallel sampling:\n- Generate N=8-32 candidate proof attempts\n- Use best-of-N selection\n- Each sample at temperature 0.6-0.8\n\n## Code Generation\n\n| Parameter | Value | Rationale |\n|-----------|-------|-----------|\n| max_tokens | 2048 | Sufficient for most functions |\n| temperature | 0.2-0.4 | Prefer deterministic output |\n\n## Creative \u002F Exploration Tasks\n\n| Parameter | Value | Rationale |\n|-----------|-------|-----------|\n| max_tokens | 4096 | Space for exploration |\n| temperature | 0.8-1.0 | Maximum creativity |\n\n## Anti-Patterns\n\n- **Too low tokens for proofs**: 512 tokens truncates chain-of-thought\n- **Too low temperature for proofs**: 0.2 misses creative tactic paths\n- **No proof plan**: Jumping to tactics without planning reduces success rate\n\n## Source Sessions\n\n- This session: APOLLO parity - increased max_tokens 512->4096, temp 0.2->0.6\n- This session: Added proof plan prompt for chain-of-thought before tactics\n",{"name":5,"description":7,"user-invocable":24},"main","70e1c350500ee750987420bbae48837afa906b7e","skills\u002Fllm-tuning-patterns","https:\u002F\u002Fgithub.com\u002Frubicanjr\u002FFinCognis","https:\u002F\u002Fgithub.com\u002Frubicanjr","linktr.ee\u002Ffinanskodu","Industrial Engineer & CIO @FinansKodu. Architecting AI-driven financial agents and autonomous workflows. 🏗️ Building systems, not just prompts.","2026-06-09 13:50:15",{"source":8,"skillName":5,"commands":37,"prompt":47,"zipAvailable":48},[38,41,44],{"tool":39,"command":40},"npx","npx skills add rubicanjr\u002FFinCognis --skill llm-tuning-patterns",{"tool":42,"command":43},"bunx","bunx skills add rubicanjr\u002FFinCognis --skill llm-tuning-patterns",{"tool":45,"command":46},"pnpm","pnpm dlx skills add rubicanjr\u002FFinCognis --skill llm-tuning-patterns","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Frubicanjr\u002FFinCognis\u002Fblob\u002Fmain\u002Fskills\u002Fllm-tuning-patterns\u002FSKILL.md",true,[],[51,72,80,99,129,148],{"skillNo":52,"name":53,"nameZh":6,"description":54,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":55,"categories":56,"keywords":62,"publishedAt":23},"01M6NWS9DAXJTADT4T5MB49X8W","workflow-router","Goal-based workflow orchestration - routes tasks to specialist agents based on user goals","https:\u002F\u002Fgithub.com\u002Frubicanjr\u002FFinCognis\u002Fblob\u002Fmain\u002Fskills\u002Fworkflow-router\u002FSKILL.md",[57],{"categoryNo":58,"categoryName":59,"taxonomyType":60,"taxonomyTypeDesc":61},"01JBSEEDCAT000000000000023","Skill与智能体开发","SCENARIO","使用场景",[63,64,65,66,67,68,69,70,71],"workflow router","workflow","router","orchestration","goals","goal","user","workflow orchestration","tasks",{"skillNo":73,"name":74,"nameZh":6,"description":75,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":76,"categories":77,"keywords":78,"publishedAt":23},"01M6NWS3RETSC0SHP16HP66E14","wiring","Wiring Verification","https:\u002F\u002Fgithub.com\u002Frubicanjr\u002FFinCognis\u002Fblob\u002Fmain\u002Fskills\u002Fwiring\u002FSKILL.md",[],[74,79,22],"verification",{"skillNo":81,"name":82,"nameZh":6,"description":83,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":84,"categories":85,"keywords":89,"publishedAt":23},"01M6NWS0XW9CVQZBDA1EGW0TWC","websocket-patterns","Connection management, room patterns, reconnection strategies, message buffering, and binary protocol design.","https:\u002F\u002Fgithub.com\u002Frubicanjr\u002FFinCognis\u002Fblob\u002Fmain\u002Fskills\u002Fwebsocket-patterns\u002Fskill.md",[86],{"categoryNo":87,"categoryName":88,"taxonomyType":60,"taxonomyTypeDesc":61},"01JBSEEDCAT000000000000015","视觉设计",[90,20,91,92,93,94,95,96,97,98],"websocket","websocket patterns","design","protocol","management","message","strategies","binary","connection",{"skillNo":100,"name":101,"nameZh":6,"description":102,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":103,"categories":104,"keywords":109,"publishedAt":23},"01M6NWRDQDX9ZNETAWRRMQWV52","visual-verdict","Screenshot comparison QA for frontend development. Takes a screenshot of the current implementation, scores it across multiple visual dimensions, and returns a structured PASS\u002FREVISE\u002FFAIL verdict with concrete fixes. Use when implementing UI from a design reference or verifying visual correctness.","https:\u002F\u002Fgithub.com\u002Frubicanjr\u002FFinCognis\u002Fblob\u002Fmain\u002Fskills\u002Fvisual-verdict\u002FSKILL.md",[105,108],{"categoryNo":106,"categoryName":107,"taxonomyType":60,"taxonomyTypeDesc":61},"01JBSEEDCAT000000000000009","前端开发",{"categoryNo":58,"categoryName":59,"taxonomyType":60,"taxonomyTypeDesc":61},[110,111,112,113,114,115,116,117,118,119,120,121,122,92,123,124,125,126,127,128],"visual","verdict","visual verdict","ui","across","revise","comparison","correctness","implementing","for","fixes","dimensions","frontend","development","qa","screenshot","returns","use","frontend development",{"skillNo":130,"name":131,"nameZh":6,"description":132,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":133,"categories":134,"keywords":135,"publishedAt":23},"01M6NWR8HSNRS7MNDK2J02QYMN","verification-loop","Comprehensive verification system covering build, types, lint, tests, security, and diff review before a PR.","https:\u002F\u002Fgithub.com\u002Frubicanjr\u002FFinCognis\u002Fblob\u002Fmain\u002Fskills\u002Fverification-loop\u002FSKILL.md",[],[136,137,79,138,139,140,141,142,143,144,145,146,147],"verification loop","loop","lint","review","security","pr","system","diff","build","tests","diff review","types",{"skillNo":149,"name":150,"nameZh":6,"description":151,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":152,"categories":153,"keywords":154,"publishedAt":23},"01M6NWR4HHSFR0VR5EXFRWA747","vector-db-patterns","Embedding strategies, ANN algorithms, hybrid search, RAG chunking strategies, and reranking for semantic search and retrieval.","https:\u002F\u002Fgithub.com\u002Frubicanjr\u002FFinCognis\u002Fblob\u002Fmain\u002Fskills\u002Fvector-db-patterns\u002Fskill.md",[],[155,156,157,158,20,159,160,161,162,96,163,164,165,166,167,119,168,169],"vector db patterns","vector db","db","vector","semantic search","search","embedding strategies","algorithms","chunking","semantic","retrieval","embedding","Hybrid Search","hybrid","rag"]