[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M79429PFS8WV29JEPRSZHESA":3,"skill-same:01M79429PFS8WV29JEPRSZHESA":66},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":15,"categories":16,"keywords":17,"publishedAt":18,"linkOnly":19,"markdownContent":20,"markdownBody":21,"frontMatter":22,"refName":26,"commitSha":27,"skillDirectory":28,"forksCount":29,"repositoryHtmlUrl":30,"repositoryHomepageUrl":6,"ownerHtmlUrl":31,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":32,"install":33,"resources":45},"01M79429PFS8WV29JEPRSZHESA","builder",null,"Use this RAGs sub-skill to build a Streamlit RAG agent from files, directories, URLs, optional web search, model settings, and RAG parameters.","VectorSpaceLab\u002FAREX-Skill","VectorSpaceLab","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F180637102?v=4","01M7538C5WPSJGWF7KHR4EFZ9F","01M7538C6R2GFP53NPXTCZNBRT",351,"MIT","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Frags\u002Fsub-skills\u002Fbuilder\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: builder\ndescription: \"Use this RAGs sub-skill to build a Streamlit RAG agent from files,\n  directories, URLs, optional web search, model settings, and RAG parameters.\"\ndisable-model-invocation: true\nmetadata:\n  disco-role: operating\nlicense: MIT\n---\n\n# RAGs Builder\n\nUse this sub-skill when the task is to create a new RAGs bot, diagnose the Home\npage builder flow, or use the builder APIs directly. RAGs is a Streamlit app\nthat asks a builder agent to create a retrieval-augmented chat agent over a\nsingle data source.\n\n## Read This When\n\n- The user wants to build a RAG bot from local files, one directory, or URLs.\n- The task mentions the Home page, builder agent, task description, system\n  prompt generation, `RAGAgentBuilder`, or `RAGParams`.\n- The user needs to choose `top_k`, `chunk_size`, embedding model, LLM provider,\n  summarization, or optional `web_search`.\n- The user asks why source loading, OpenAI secrets, web search, or multimodal\n  setup fails before the agent exists.\n\nIf the bot already exists and the user wants to edit or delete it, read\n[`..\u002Fconfiguration\u002FSKILL.md`](..\u002Fconfiguration\u002FSKILL.md). If the user wants to\nask questions to an existing bot, read [`..\u002Fchat\u002FSKILL.md`](..\u002Fchat\u002FSKILL.md).\n\n## Core Workflow\n\n1. Confirm dependencies and secrets first. The app expects a Streamlit secret\n   named `openai_key`; provider-specific keys are needed only for their routes.\n2. Validate the data-source choice with\n   [`scripts\u002Fvalidate_source_selection.py`](scripts\u002Fvalidate_source_selection.py)\n   before invoking builder logic. RAGs accepts exactly one source kind: local\n   files, one directory, or URLs.\n3. Use the builder conversation to collect the task description and let the\n   builder create the system prompt.\n4. Load data, inspect or set RAG parameters, optionally add `web_search` when a\n   `metaphor_key` secret is configured, then create the agent.\n5. After creation, route to the configuration and chat sub-skills for later\n   editing and querying.\n\nFor detailed step-by-step routes, model identifiers, data-source rules, and\nvalidation checks, read [`references\u002Fworkflows.md`](references\u002Fworkflows.md).\nFor signatures and defaults verified from the package inspection environment,\nread [`references\u002Fapi-reference.md`](references\u002Fapi-reference.md).\nFor failure symptoms and fixes, read\n[`references\u002Ftroubleshooting.md`](references\u002Ftroubleshooting.md).\n\n## Important Operating Facts\n\n- `RAGParams` defaults are `include_summarization=False`, `top_k=2`,\n  `chunk_size=1024`, `embed_model=\"default\"`, and\n  `llm=\"gpt-4-1106-preview\"`.\n- `load_data` requires exactly one of `file_names`, `directory`, or `urls` and\n  raises a `ValueError` when zero or multiple source kinds are supplied.\n- Unprefixed LLM names are treated as OpenAI models. Recognized prefixes are\n  `openai:`, `anthropic:`, `replicate:`, and `local:`.\n- The web-search tool is only exposed when a `metaphor_key` Streamlit secret is\n  present, and its tool name is `web_search`.\n- The beta multimodal builder path is optional. The minimum verified\n  environment imported multimodal classes but did not install the torch,\n  torchvision, or CLIP runtime stack and did not run external model calls.\n\n## Boundaries\n\nDo not claim that a full OpenAI, Anthropic, Replicate, Metaphor, URL-loading, or\nmultimodal chat call was verified unless the user supplies credentials\u002Fnetwork\nand asks for that execution. For safe checks, use the bundled validator and the\nroot installation diagnostic instead of launching a long-running Streamlit\nserver.\n","# RAGs Builder\n\nUse this sub-skill when the task is to create a new RAGs bot, diagnose the Home\npage builder flow, or use the builder APIs directly. RAGs is a Streamlit app\nthat asks a builder agent to create a retrieval-augmented chat agent over a\nsingle data source.\n\n## Read This When\n\n- The user wants to build a RAG bot from local files, one directory, or URLs.\n- The task mentions the Home page, builder agent, task description, system\n  prompt generation, `RAGAgentBuilder`, or `RAGParams`.\n- The user needs to choose `top_k`, `chunk_size`, embedding model, LLM provider,\n  summarization, or optional `web_search`.\n- The user asks why source loading, OpenAI secrets, web search, or multimodal\n  setup fails before the agent exists.\n\nIf the bot already exists and the user wants to edit or delete it, read\n[`..\u002Fconfiguration\u002FSKILL.md`](..\u002Fconfiguration\u002FSKILL.md). If the user wants to\nask questions to an existing bot, read [`..\u002Fchat\u002FSKILL.md`](..\u002Fchat\u002FSKILL.md).\n\n## Core Workflow\n\n1. Confirm dependencies and secrets first. The app expects a Streamlit secret\n   named `openai_key`; provider-specific keys are needed only for their routes.\n2. Validate the data-source choice with\n   [`scripts\u002Fvalidate_source_selection.py`](scripts\u002Fvalidate_source_selection.py)\n   before invoking builder logic. RAGs accepts exactly one source kind: local\n   files, one directory, or URLs.\n3. Use the builder conversation to collect the task description and let the\n   builder create the system prompt.\n4. Load data, inspect or set RAG parameters, optionally add `web_search` when a\n   `metaphor_key` secret is configured, then create the agent.\n5. After creation, route to the configuration and chat sub-skills for later\n   editing and querying.\n\nFor detailed step-by-step routes, model identifiers, data-source rules, and\nvalidation checks, read [`references\u002Fworkflows.md`](references\u002Fworkflows.md).\nFor signatures and defaults verified from the package inspection environment,\nread [`references\u002Fapi-reference.md`](references\u002Fapi-reference.md).\nFor failure symptoms and fixes, read\n[`references\u002Ftroubleshooting.md`](references\u002Ftroubleshooting.md).\n\n## Important Operating Facts\n\n- `RAGParams` defaults are `include_summarization=False`, `top_k=2`,\n  `chunk_size=1024`, `embed_model=\"default\"`, and\n  `llm=\"gpt-4-1106-preview\"`.\n- `load_data` requires exactly one of `file_names`, `directory`, or `urls` and\n  raises a `ValueError` when zero or multiple source kinds are supplied.\n- Unprefixed LLM names are treated as OpenAI models. Recognized prefixes are\n  `openai:`, `anthropic:`, `replicate:`, and `local:`.\n- The web-search tool is only exposed when a `metaphor_key` Streamlit secret is\n  present, and its tool name is `web_search`.\n- The beta multimodal builder path is optional. The minimum verified\n  environment imported multimodal classes but did not install the torch,\n  torchvision, or CLIP runtime stack and did not run external model calls.\n\n## Boundaries\n\nDo not claim that a full OpenAI, Anthropic, Replicate, Metaphor, URL-loading, or\nmultimodal chat call was verified unless the user supplies credentials\u002Fnetwork\nand asks for that execution. For safe checks, use the bundled validator and the\nroot installation diagnostic instead of launching a long-running Streamlit\nserver.\n",{"name":5,"license":14,"metadata":23,"description":7,"disable-model-invocation":25},{"disco-role":24},"operating",true,"main","ac3fe1afa80fb9a09775ecfb2b6cc3ba850a2db6","skills\u002Frepositories\u002Frepo-skills\u002Frags\u002Fsub-skills\u002Fbuilder",27,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab","2026-09-03 10:25:21",{"source":8,"skillName":5,"commands":34,"prompt":44,"zipAvailable":25},[35,38,41],{"tool":36,"command":37},"npx","npx skills add VectorSpaceLab\u002FAREX-Skill --skill builder",{"tool":39,"command":40},"bunx","bunx skills add VectorSpaceLab\u002FAREX-Skill --skill builder",{"tool":42,"command":43},"pnpm","pnpm dlx skills add VectorSpaceLab\u002FAREX-Skill --skill builder","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Frags\u002Fsub-skills\u002Fbuilder\u002FSKILL.md",[46,52,56,60],{"path":47,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":50,"sourceHtmlUrl":51},"references\u002Fapi-reference.md","REFERENCE","参考文档",4264,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Frags\u002Fsub-skills\u002Fbuilder\u002Freferences\u002Fapi-reference.md",{"path":53,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":54,"sourceHtmlUrl":55},"references\u002Ftroubleshooting.md",3621,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Frags\u002Fsub-skills\u002Fbuilder\u002Freferences\u002Ftroubleshooting.md",{"path":57,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":58,"sourceHtmlUrl":59},"references\u002Fworkflows.md",4650,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Frags\u002Fsub-skills\u002Fbuilder\u002Freferences\u002Fworkflows.md",{"path":61,"resourceType":62,"resourceTypeDesc":63,"sizeBytes":64,"sourceHtmlUrl":65},"scripts\u002Fvalidate_source_selection.py","SCRIPT","脚本",3799,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Frags\u002Fsub-skills\u002Fbuilder\u002Fscripts\u002Fvalidate_source_selection.py",[67,75,82,89,96,103],{"skillNo":68,"name":69,"nameZh":6,"description":70,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":71,"licenseFileUrl":6,"sourceHtmlUrl":72,"categories":73,"keywords":74,"publishedAt":18},"01M799NYSH1R3RVAHSM2D88Y84","passnet-triton-opt","Make a single PassNet Triton kernel fast AND numerically faithful: performance model (when a replacement can win at all), block\u002Fgrid\u002Fwarp tuning, autotune policy, launch overhead, and per-op numeric recipes to pass the dtype baseline tolerances. Use when a pass matches and is correct but speedup below expected, or when correctness fails by small numeric margins.","Apache-2.0","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Ftask-oriented\u002FPassNet\u002Fskills\u002Fpassnet-triton-opt\u002FSKILL.md",[],[],{"skillNo":76,"name":77,"nameZh":6,"description":78,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":71,"licenseFileUrl":6,"sourceHtmlUrl":79,"categories":80,"keywords":81,"publishedAt":18},"01M799NS5APEC0N1FZ7XZVW6BN","passnet-solve","END-TO-END playbook for solving one PassNet sample: analyze the computation graphs, decide the optimization strategy (what to fuse, what to replace, what to leave alone), drive the iteration loop, and maximize the sample score. This is the ENTRY skill — invoke it first for any PassNet optimization task; it tells you when to use passnet-pattern-fusion, passnet-triton-opt and passnet-feedback.","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Ftask-oriented\u002FPassNet\u002Fskills\u002Fpassnet-solve\u002FSKILL.md",[],[],{"skillNo":83,"name":84,"nameZh":6,"description":85,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":71,"licenseFileUrl":6,"sourceHtmlUrl":86,"categories":87,"keywords":88,"publishedAt":18},"01M799NNY9BKSM21V5R7QVTBJX","passnet-skill","PassNet GPU kernel optimization via compiler passes. Design and implement Triton-based optimization passes, create pass files under .\u002Fpass_dir\u002F, self-evaluate with pass_evaluator, and iterate to maximize GPU speedup.","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Ftask-oriented\u002FPassNet\u002Fskills\u002Fpassnet-skill\u002FSKILL.md",[],[],{"skillNo":90,"name":91,"nameZh":6,"description":92,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":71,"licenseFileUrl":6,"sourceHtmlUrl":93,"categories":94,"keywords":95,"publishedAt":18},"01M799NHH6NP7Z1RK77KGXAN6S","passnet-pattern-fusion","Author PassNet pass files: write patterns that actually MATCH the FX graph, pick fusion regions (which ops to absorb into one Triton kernel), and structure multi-pass submissions with the mandatory shared-dispatch architecture. Use when creating or fixing pattern()\u002Freplacement_args()\u002Freplacement_func() files, when a pass \"failed to match\", or when deciding how to fuse multiple kernels\u002Fops.","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Ftask-oriented\u002FPassNet\u002Fskills\u002Fpassnet-pattern-fusion\u002FSKILL.md",[],[],{"skillNo":97,"name":98,"nameZh":6,"description":99,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":71,"licenseFileUrl":6,"sourceHtmlUrl":100,"categories":101,"keywords":102,"publishedAt":18},"01M799ND2J4WNC2PGAZ0A06NDF","passnet-orchestrate","Round\u002Fbatch planning and decision support for PassNet graph optimization. Use for multi-sample triage, worker allocation, eval-budget policy, and post-eval keep\u002Frevert decisions. For solving one concrete sample end to end, use passnet-solve as the entry skill; consult this skill only when you need broader planning or a specific decision gate.","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Ftask-oriented\u002FPassNet\u002Fskills\u002Fpassnet-orchestrate\u002FSKILL.md",[],[],{"skillNo":104,"name":105,"nameZh":6,"description":106,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":71,"licenseFileUrl":6,"sourceHtmlUrl":107,"categories":108,"keywords":109,"publishedAt":18},"01M799N8SBTMY02ZDG2AZF5ZWE","passnet-feedback","Fast, deterministic feedback for PassNet work: pre-flight pattern\u002Fmatch verification WITHOUT burning a GPU evaluation, per-node bottleneck analysis of a sample's graphs, and eval-log parsing into per-variant status + estimated score + failure classification. Use BEFORE every GPU evaluation (check_pattern), at the START of a sample (analyze_graph), and AFTER every evaluation (parse_eval_log).","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Ftask-oriented\u002FPassNet\u002Fskills\u002Fpassnet-feedback\u002FSKILL.md",[],[]]