[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M78YZD373RASWN63XBPTYEB5":3,"skill-same:01M78YZD373RASWN63XBPTYEB5":67},{"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":27,"commitSha":28,"skillDirectory":29,"forksCount":30,"repositoryHtmlUrl":31,"repositoryHomepageUrl":6,"ownerHtmlUrl":32,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":33,"install":34,"resources":46},"01M78YZD373RASWN63XBPTYEB5","providers",null,"Choose, configure, inspect, and troubleshoot LangExtract model providers and batch backends.","VectorSpaceLab\u002FAREX-Skill","VectorSpaceLab","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F180637102?v=4","01M7538C5WPSJGWF7KHR4EFZ9F","01M7538C6R2GFP53NPXTCZNBRT",351,"Apache-2.0","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Flangextract\u002Fsub-skills\u002Fproviders\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: providers\ndescription: \"Choose, configure, inspect, and troubleshoot LangExtract model\n  providers and batch backends.\"\nmetadata:\n  disco-role: operating\ndisable-model-invocation: true\nlicense: Apache 2.0\n---\n\n# LangExtract Providers\n\nUse this sub-skill when a task depends on selecting or configuring the LangExtract language-model backend: Gemini, Vertex AI, OpenAI\u002FGPT, OpenAI-compatible endpoints, Ollama\u002Flocal models, provider routing, provider kwargs, or batch behavior.\n\n## Fast route\n\n1. Decide whether the user is doing provider setup or prompt\u002Fextraction design.\n   - Provider\u002Fbackend choice, API keys, `ModelConfig`, `language_model_params`, batch mode, Ollama daemon\u002Fmodel checks: stay here.\n   - Prompt, examples, `output_schema` shape, tokenization, resolver tuning, or extraction result grounding: route to [extraction](..\u002Fextraction\u002FSKILL.md).\n   - Custom provider package scaffolding, entry points, plugin conflicts, or schema implementation for a new provider: route to [provider-plugins](..\u002Fprovider-plugins\u002FSKILL.md).\n   - JSONL persistence or HTML output: route to [visualization](..\u002Fvisualization\u002FSKILL.md).\n2. Inspect routing without secrets or network by running the bundled route checker:\n   ```bash\n   python sub-skills\u002Fproviders\u002Fscripts\u002Fcheck_provider_routes.py\n   ```\n3. Use [references\u002Fproviders.md](references\u002Fproviders.md) for exact provider patterns, factory APIs, environment-variable precedence, schema\u002Ffence interactions, and batch configuration.\n4. Use [references\u002Ftroubleshooting.md](references\u002Ftroubleshooting.md) when provider creation or inference fails.\n5. For local Ollama, start with a preflight-only check before inference:\n   ```bash\n   python sub-skills\u002Fproviders\u002Fscripts\u002Follama_demo.py --model gemma2:2b --preflight-only\n   ```\n\n## Provider selection checklist\n\n- Prefer `model_id` auto-routing for known IDs: `gemini...` for Gemini, `gpt-4...` or `gpt-5...` for OpenAI, and common local IDs such as `gemma2:2b`, `llama3.2:1b`, `qwen...`, or `gpt-oss...` for Ollama.\n- Use `langextract.factory.ModelConfig(provider=..., provider_kwargs=...)` when a model ID does not match a built-in pattern, when an OpenAI-compatible endpoint uses a non-GPT ID, or when multiple providers could match.\n- Do not put secrets in generated code. Read API keys from environment variables or caller-owned secret stores.\n- Treat live Gemini\u002FOpenAI\u002FVertex\u002FOllama checks as optional. This sub-skill only establishes configuration and local\u002Fno-network inspection unless the caller explicitly supplies credentials or a local daemon.\n- Batch APIs are opt-in and only trigger when the prompt\u002Fchunk count meets the configured threshold.\n\n## Runtime helpers\n\n- `scripts\u002Fcheck_provider_routes.py`: safe, no-network provider registry and sample model-ID resolver.\n- `scripts\u002Follama_demo.py`: local Ollama preflight plus optional tiny extraction demo. It does not require the source repository checkout.\n\n## Evidence basis\n\nThis guidance is distilled from `README.md`, `langextract\u002Fproviders\u002FREADME.md`, `docs\u002Fexamples\u002Fbatch_api_example.md`, `examples\u002Follama\u002F`, `langextract\u002Ffactory.py`, `langextract\u002Fproviders\u002Frouter.py`, `builtin_registry.py`, `patterns.py`, `gemini.py`, `openai.py`, `ollama.py`, `gemini_batch.py`, `openai_batch.py`, and provider schema modules. Related native verification candidates include `factory_test.py`, `factory_schema_test.py`, `provider_schema_test.py`, `inference_test.py`, `test_kwargs_passthrough.py`, `test_gemini_batch_api.py`, `gemini_retry_test.py`, `openai_batch_test.py`, optional `test_live_api.py`, and optional `test_ollama_integration.py`; do not run those from this sub-skill unless the root verification plan asks for them.\n","# LangExtract Providers\n\nUse this sub-skill when a task depends on selecting or configuring the LangExtract language-model backend: Gemini, Vertex AI, OpenAI\u002FGPT, OpenAI-compatible endpoints, Ollama\u002Flocal models, provider routing, provider kwargs, or batch behavior.\n\n## Fast route\n\n1. Decide whether the user is doing provider setup or prompt\u002Fextraction design.\n   - Provider\u002Fbackend choice, API keys, `ModelConfig`, `language_model_params`, batch mode, Ollama daemon\u002Fmodel checks: stay here.\n   - Prompt, examples, `output_schema` shape, tokenization, resolver tuning, or extraction result grounding: route to [extraction](..\u002Fextraction\u002FSKILL.md).\n   - Custom provider package scaffolding, entry points, plugin conflicts, or schema implementation for a new provider: route to [provider-plugins](..\u002Fprovider-plugins\u002FSKILL.md).\n   - JSONL persistence or HTML output: route to [visualization](..\u002Fvisualization\u002FSKILL.md).\n2. Inspect routing without secrets or network by running the bundled route checker:\n   ```bash\n   python sub-skills\u002Fproviders\u002Fscripts\u002Fcheck_provider_routes.py\n   ```\n3. Use [references\u002Fproviders.md](references\u002Fproviders.md) for exact provider patterns, factory APIs, environment-variable precedence, schema\u002Ffence interactions, and batch configuration.\n4. Use [references\u002Ftroubleshooting.md](references\u002Ftroubleshooting.md) when provider creation or inference fails.\n5. For local Ollama, start with a preflight-only check before inference:\n   ```bash\n   python sub-skills\u002Fproviders\u002Fscripts\u002Follama_demo.py --model gemma2:2b --preflight-only\n   ```\n\n## Provider selection checklist\n\n- Prefer `model_id` auto-routing for known IDs: `gemini...` for Gemini, `gpt-4...` or `gpt-5...` for OpenAI, and common local IDs such as `gemma2:2b`, `llama3.2:1b`, `qwen...`, or `gpt-oss...` for Ollama.\n- Use `langextract.factory.ModelConfig(provider=..., provider_kwargs=...)` when a model ID does not match a built-in pattern, when an OpenAI-compatible endpoint uses a non-GPT ID, or when multiple providers could match.\n- Do not put secrets in generated code. Read API keys from environment variables or caller-owned secret stores.\n- Treat live Gemini\u002FOpenAI\u002FVertex\u002FOllama checks as optional. This sub-skill only establishes configuration and local\u002Fno-network inspection unless the caller explicitly supplies credentials or a local daemon.\n- Batch APIs are opt-in and only trigger when the prompt\u002Fchunk count meets the configured threshold.\n\n## Runtime helpers\n\n- `scripts\u002Fcheck_provider_routes.py`: safe, no-network provider registry and sample model-ID resolver.\n- `scripts\u002Follama_demo.py`: local Ollama preflight plus optional tiny extraction demo. It does not require the source repository checkout.\n\n## Evidence basis\n\nThis guidance is distilled from `README.md`, `langextract\u002Fproviders\u002FREADME.md`, `docs\u002Fexamples\u002Fbatch_api_example.md`, `examples\u002Follama\u002F`, `langextract\u002Ffactory.py`, `langextract\u002Fproviders\u002Frouter.py`, `builtin_registry.py`, `patterns.py`, `gemini.py`, `openai.py`, `ollama.py`, `gemini_batch.py`, `openai_batch.py`, and provider schema modules. Related native verification candidates include `factory_test.py`, `factory_schema_test.py`, `provider_schema_test.py`, `inference_test.py`, `test_kwargs_passthrough.py`, `test_gemini_batch_api.py`, `gemini_retry_test.py`, `openai_batch_test.py`, optional `test_live_api.py`, and optional `test_ollama_integration.py`; do not run those from this sub-skill unless the root verification plan asks for them.\n",{"name":5,"license":23,"metadata":24,"description":7,"disable-model-invocation":26},"Apache 2.0",{"disco-role":25},"operating",true,"main","ac3fe1afa80fb9a09775ecfb2b6cc3ba850a2db6","skills\u002Frepositories\u002Frepo-skills\u002Flangextract\u002Fsub-skills\u002Fproviders",27,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab","2026-09-03 10:25:21",{"source":8,"skillName":5,"commands":35,"prompt":45,"zipAvailable":26},[36,39,42],{"tool":37,"command":38},"npx","npx skills add VectorSpaceLab\u002FAREX-Skill --skill providers",{"tool":40,"command":41},"bunx","bunx skills add VectorSpaceLab\u002FAREX-Skill --skill providers",{"tool":43,"command":44},"pnpm","pnpm dlx skills add VectorSpaceLab\u002FAREX-Skill --skill providers","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Flangextract\u002Fsub-skills\u002Fproviders\u002FSKILL.md",[47,53,57,63],{"path":48,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":51,"sourceHtmlUrl":52},"references\u002Fproviders.md","REFERENCE","参考文档",10879,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Flangextract\u002Fsub-skills\u002Fproviders\u002Freferences\u002Fproviders.md",{"path":54,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":55,"sourceHtmlUrl":56},"references\u002Ftroubleshooting.md",7797,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Flangextract\u002Fsub-skills\u002Fproviders\u002Freferences\u002Ftroubleshooting.md",{"path":58,"resourceType":59,"resourceTypeDesc":60,"sizeBytes":61,"sourceHtmlUrl":62},"scripts\u002Fcheck_provider_routes.py","SCRIPT","脚本",5117,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Flangextract\u002Fsub-skills\u002Fproviders\u002Fscripts\u002Fcheck_provider_routes.py",{"path":64,"resourceType":59,"resourceTypeDesc":60,"sizeBytes":65,"sourceHtmlUrl":66},"scripts\u002Follama_demo.py",5638,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Flangextract\u002Fsub-skills\u002Fproviders\u002Fscripts\u002Follama_demo.py",[68,75,82,89,96,103],{"skillNo":69,"name":70,"nameZh":6,"description":71,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"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.","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":14,"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":14,"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":14,"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":14,"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":14,"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",[],[]]