[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M78YYKE5ZM59VSSEE8P6RZGS":3,"skill-same:01M78YYKE5ZM59VSSEE8P6RZGS":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},"01M78YYKE5ZM59VSSEE8P6RZGS","agents-and-middleware",null,"Work on the actively maintained LangChain v1 agent package: init_chat_model, create_agent, structured output, tools, middleware, embeddings initialization, provider routing, and agent runtime customization. Use for libs\u002Flangchain_v1 agent workflows and route low-level core primitives or provider implementation details to sibling skills.","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\u002Flangchain\u002Fsub-skills\u002Fagents-and-middleware\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: agents-and-middleware\ndescription: \"Work on the actively maintained LangChain v1 agent package:\n  init_chat_model, create_agent, structured output, tools, middleware,\n  embeddings initialization, provider routing, and agent runtime customization.\n  Use for libs\u002Flangchain_v1 agent workflows and route low-level core primitives\n  or provider implementation details to sibling skills.\"\ndisable-model-invocation: true\nmetadata:\n  disco-role: operating\nlicense: MIT\n---\n\n# Agents and Middleware\n\nUse this sub-skill for practical work in the actively maintained `langchain` package around v1 agents, middleware, model initialization, structured output, tools, and embeddings. This guidance is distilled from the LangChain v1 source and tests and is self-contained for future coding agents.\n\n## When to Use\n\n- User asks about `init_chat_model`, `create_agent`, `response_format`, tool schemas, injected tool runtime, or configurable chat models.\n- User asks to add, configure, or debug middleware such as retry, fallback, human-in-the-loop, summarization, tool\u002Fmodel call limits, PII redaction, file search, shell execution, todo planning, provider tool search, tool retry, or tool selection.\n- User asks why provider resolution fails for a chat or embedding model, especially when optional integration packages or credentials are missing.\n- User asks to customize agent state, runtime context, checkpointer\u002Fstore usage, interrupts, streaming, or debug behavior at the `langchain` v1 layer.\n\n## Route Elsewhere\n\n- For low-level runnable, message, tool, callback, language-model, or embedding primitives from `langchain_core`, use `..\u002Fcore-primitives\u002FSKILL.md`.\n- For provider implementation packages such as `langchain-openai`, `langchain-anthropic`, `langchain-ollama`, or provider-specific parameters\u002Fclasses, use `..\u002Fintegrations\u002FSKILL.md`.\n- For legacy `langchain-classic` imports, chains, retrievers, or classic agents, use the sibling skill that owns classic APIs instead of rewriting v1 guidance.\n\n## Reference Map\n\n- Start with [references\u002Fagent-workflows.md](references\u002Fagent-workflows.md) for source layout, import paths, common edit workflows, validation commands, structured output, tools, embeddings, and provider initialization.\n- Use [references\u002Fmiddleware-reference.md](references\u002Fmiddleware-reference.md) for middleware families, exported classes, hook styles, ordering, state\u002Fcontext patterns, and safety constraints.\n- Use [references\u002Ftroubleshooting.md](references\u002Ftroubleshooting.md) for missing provider packages, credentials\u002Fnetwork skips, structured output\u002Ftool validation, middleware ordering, HITL\u002Fshell\u002Ffile-search safety, and v1-vs-classic confusion.\n- Run [scripts\u002Fagent_import_smoke.py](scripts\u002Fagent_import_smoke.py) as a safe import-only smoke check when an environment is available.\n\n## Fast Workflow\n\n1. Confirm the target package is `libs\u002Flangchain_v1`; its distribution name is `langchain` and its package imports are `langchain.*`.\n2. Inspect the nearest v1 public API files first: `langchain\u002Fchat_models\u002Fbase.py`, `langchain\u002Fagents\u002Ffactory.py`, `langchain\u002Fagents\u002Fstructured_output.py`, `langchain\u002Fagents\u002Fmiddleware\u002F`, `langchain\u002Ftools\u002F`, and `langchain\u002Fembeddings\u002Fbase.py`.\n3. Prefer tests under `tests\u002Funit_tests\u002Fagents`, `tests\u002Funit_tests\u002Fchat_models`, `tests\u002Funit_tests\u002Ftools`, and `tests\u002Funit_tests\u002Fembeddings` for expected public behavior.\n4. For package validation, use `uv` from `libs\u002Flangchain_v1`; do not use `pip`, `poetry`, or `conda` directly for this monorepo.\n5. Skip network-backed model invocations unless credentials, provider packages, and user permission are present; use fake models or import checks for local validation.\n\n## Safe Validation\n\nFrom `libs\u002Flangchain_v1`, use targeted package tests when `uv` is available:\n\n```bash\nuv run --group test pytest tests\u002Funit_tests\u002Fchat_models\u002Ftest_chat_models.py tests\u002Funit_tests\u002Fagents\u002Ftest_response_format.py tests\u002Funit_tests\u002Ftools\u002Ftest_imports.py tests\u002Funit_tests\u002Fembeddings\u002Ftest_base.py\n```\n\nFrom this sub-skill directory, use the bundled smoke script with any Python environment that already has `langchain` installed:\n\n```bash\npython scripts\u002Fagent_import_smoke.py\n```\n\nThe smoke script imports public APIs only and does not call providers, networks, shell commands, or external files.\n\n## Guardrails\n\n- Keep v1 import paths explicit: `langchain.chat_models`, `langchain.agents`, `langchain.agents.middleware`, `langchain.tools`, and `langchain.embeddings`.\n- Do not add provider-specific hard dependencies to `langchain` core code unless the package metadata intentionally lists them as optional extras.\n- Do not run shell\u002Ffile-search\u002FHITL examples as unattended validation; these require user-reviewed safety decisions.\n- Do not link future runtime instructions to source-checkout docs, examples, tests, or absolute local paths; distill facts into this sub-skill or bundled references.\n","# Agents and Middleware\n\nUse this sub-skill for practical work in the actively maintained `langchain` package around v1 agents, middleware, model initialization, structured output, tools, and embeddings. This guidance is distilled from the LangChain v1 source and tests and is self-contained for future coding agents.\n\n## When to Use\n\n- User asks about `init_chat_model`, `create_agent`, `response_format`, tool schemas, injected tool runtime, or configurable chat models.\n- User asks to add, configure, or debug middleware such as retry, fallback, human-in-the-loop, summarization, tool\u002Fmodel call limits, PII redaction, file search, shell execution, todo planning, provider tool search, tool retry, or tool selection.\n- User asks why provider resolution fails for a chat or embedding model, especially when optional integration packages or credentials are missing.\n- User asks to customize agent state, runtime context, checkpointer\u002Fstore usage, interrupts, streaming, or debug behavior at the `langchain` v1 layer.\n\n## Route Elsewhere\n\n- For low-level runnable, message, tool, callback, language-model, or embedding primitives from `langchain_core`, use `..\u002Fcore-primitives\u002FSKILL.md`.\n- For provider implementation packages such as `langchain-openai`, `langchain-anthropic`, `langchain-ollama`, or provider-specific parameters\u002Fclasses, use `..\u002Fintegrations\u002FSKILL.md`.\n- For legacy `langchain-classic` imports, chains, retrievers, or classic agents, use the sibling skill that owns classic APIs instead of rewriting v1 guidance.\n\n## Reference Map\n\n- Start with [references\u002Fagent-workflows.md](references\u002Fagent-workflows.md) for source layout, import paths, common edit workflows, validation commands, structured output, tools, embeddings, and provider initialization.\n- Use [references\u002Fmiddleware-reference.md](references\u002Fmiddleware-reference.md) for middleware families, exported classes, hook styles, ordering, state\u002Fcontext patterns, and safety constraints.\n- Use [references\u002Ftroubleshooting.md](references\u002Ftroubleshooting.md) for missing provider packages, credentials\u002Fnetwork skips, structured output\u002Ftool validation, middleware ordering, HITL\u002Fshell\u002Ffile-search safety, and v1-vs-classic confusion.\n- Run [scripts\u002Fagent_import_smoke.py](scripts\u002Fagent_import_smoke.py) as a safe import-only smoke check when an environment is available.\n\n## Fast Workflow\n\n1. Confirm the target package is `libs\u002Flangchain_v1`; its distribution name is `langchain` and its package imports are `langchain.*`.\n2. Inspect the nearest v1 public API files first: `langchain\u002Fchat_models\u002Fbase.py`, `langchain\u002Fagents\u002Ffactory.py`, `langchain\u002Fagents\u002Fstructured_output.py`, `langchain\u002Fagents\u002Fmiddleware\u002F`, `langchain\u002Ftools\u002F`, and `langchain\u002Fembeddings\u002Fbase.py`.\n3. Prefer tests under `tests\u002Funit_tests\u002Fagents`, `tests\u002Funit_tests\u002Fchat_models`, `tests\u002Funit_tests\u002Ftools`, and `tests\u002Funit_tests\u002Fembeddings` for expected public behavior.\n4. For package validation, use `uv` from `libs\u002Flangchain_v1`; do not use `pip`, `poetry`, or `conda` directly for this monorepo.\n5. Skip network-backed model invocations unless credentials, provider packages, and user permission are present; use fake models or import checks for local validation.\n\n## Safe Validation\n\nFrom `libs\u002Flangchain_v1`, use targeted package tests when `uv` is available:\n\n```bash\nuv run --group test pytest tests\u002Funit_tests\u002Fchat_models\u002Ftest_chat_models.py tests\u002Funit_tests\u002Fagents\u002Ftest_response_format.py tests\u002Funit_tests\u002Ftools\u002Ftest_imports.py tests\u002Funit_tests\u002Fembeddings\u002Ftest_base.py\n```\n\nFrom this sub-skill directory, use the bundled smoke script with any Python environment that already has `langchain` installed:\n\n```bash\npython scripts\u002Fagent_import_smoke.py\n```\n\nThe smoke script imports public APIs only and does not call providers, networks, shell commands, or external files.\n\n## Guardrails\n\n- Keep v1 import paths explicit: `langchain.chat_models`, `langchain.agents`, `langchain.agents.middleware`, `langchain.tools`, and `langchain.embeddings`.\n- Do not add provider-specific hard dependencies to `langchain` core code unless the package metadata intentionally lists them as optional extras.\n- Do not run shell\u002Ffile-search\u002FHITL examples as unattended validation; these require user-reviewed safety decisions.\n- Do not link future runtime instructions to source-checkout docs, examples, tests, or absolute local paths; distill facts into this sub-skill or bundled references.\n",{"name":5,"license":14,"metadata":23,"description":7,"disable-model-invocation":25},{"disco-role":24},"operating",true,"main","ac3fe1afa80fb9a09775ecfb2b6cc3ba850a2db6","skills\u002Frepositories\u002Frepo-skills\u002Flangchain\u002Fsub-skills\u002Fagents-and-middleware",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 agents-and-middleware",{"tool":39,"command":40},"bunx","bunx skills add VectorSpaceLab\u002FAREX-Skill --skill agents-and-middleware",{"tool":42,"command":43},"pnpm","pnpm dlx skills add VectorSpaceLab\u002FAREX-Skill --skill agents-and-middleware","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Flangchain\u002Fsub-skills\u002Fagents-and-middleware\u002FSKILL.md",[46,52,56,60],{"path":47,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":50,"sourceHtmlUrl":51},"references\u002Fagent-workflows.md","REFERENCE","参考文档",12428,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Flangchain\u002Fsub-skills\u002Fagents-and-middleware\u002Freferences\u002Fagent-workflows.md",{"path":53,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":54,"sourceHtmlUrl":55},"references\u002Fmiddleware-reference.md",11421,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Flangchain\u002Fsub-skills\u002Fagents-and-middleware\u002Freferences\u002Fmiddleware-reference.md",{"path":57,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":58,"sourceHtmlUrl":59},"references\u002Ftroubleshooting.md",10434,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Flangchain\u002Fsub-skills\u002Fagents-and-middleware\u002Freferences\u002Ftroubleshooting.md",{"path":61,"resourceType":62,"resourceTypeDesc":63,"sizeBytes":64,"sourceHtmlUrl":65},"scripts\u002Fagent_import_smoke.py","SCRIPT","脚本",2803,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Flangchain\u002Fsub-skills\u002Fagents-and-middleware\u002Fscripts\u002Fagent_import_smoke.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",[],[]]