[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7F3TYJDY1SXV3BCFF7X8Z14":3,"skill-same:01M7F3TYJDY1SXV3BCFF7X8Z14":55},{"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":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":28,"commitSha":29,"skillDirectory":30,"forksCount":31,"repositoryHtmlUrl":32,"repositoryHomepageUrl":6,"ownerHtmlUrl":33,"ownerBlogUrl":6,"ownerDescription":34,"updatedAt":35,"install":36,"resources":48},"01M7F3TYJDY1SXV3BCFF7X8Z14","llm-finetune",null,"Operate RD-Agent FT-Agent workflows for benchmark-driven LLM data processing, fine-tuning, merge, evaluation, and feedback loops.","lza6\u002FArmor-piercing-system-prompts","lza6","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F128880206?v=4","01M7EB3K7WPX5T28EJ2NVHE6EG","Chinese-tingfeng","01M7EB3K97E1KQ16QWKJMFNGX8",11,"MIT","https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F05-技能与代理开发\u002Frd-agent\u002Fsub-skills\u002Fllm-finetune\u002FSKILL.md",[],[],"2026-11-16 10:05:34",false,"---\nname: llm-finetune\ndescription: \"Operate RD-Agent FT-Agent workflows for benchmark-driven LLM data\n  processing, fine-tuning, merge, evaluation, and feedback loops.\"\nmetadata:\n  disco-role: operating\n  parent-skill: rd-agent\nlicense: MIT\ndisable-model-invocation: true\n---\n\n# RD-Agent LLM fine-tuning\n\nUse this sub-skill for `rdagent llm_finetune`, FT-Agent datasets\u002Fbenchmarks, LLaMA-Factory training configuration, OpenCompass evaluation, batch jobs, and the fine-tune UI.\n\n## Prerequisites and cost gate\n\nA realistic run can download models and datasets, call an LLM provider, start Docker or conda training environments, and consume GPU hours. Before running, confirm Linux, a usable Docker\u002Fconda backend, NVIDIA\u002FCUDA capacity when required, Hugging Face access, provider credentials, and a stable `FT_FILE_PATH`. A missing prerequisite is a blocked run, not a reason to substitute a toy score.\n\nKeep the main settings explicit and secret-free in the shell or a private `.env`:\n\n```text\nBACKEND\nCHAT_MODEL\nOPENAI_API_KEY \u002F OPENAI_API_BASE\nEMBEDDING_MODEL\nFT_FILE_PATH\nFT_Coder_CoSTEER_env_type=docker|conda\nFT_TARGET_BENCHMARK\nFT_BENCHMARK_DESCRIPTION\nFT_BASE_MODEL\nFT_UPPER_DATA_SIZE_LIMIT\nFT_API_MAX_WORKERS\nFT_STRONG_MODELS \u002F FT_WEAK_MODELS\nHF_TOKEN (only when required)\n```\n\nUse [validate_ft_env.py](scripts\u002Fvalidate_ft_env.py) to print presence only; it never prints values.\n\n## Single-task workflow\n\nStart with help and a bounded configuration:\n\n```bash\nrdagent llm_finetune --help\nrdagent llm_finetune \\\n  --benchmark \u003Cbenchmark> \\\n  --benchmark-description \"\u003Ctask and output format>\" \\\n  --base-model \u003Chuggingface-model-id> \\\n  --loop-n 1 \\\n  --timeout \u003Cshort-budget>\n```\n\nThe checked-in implementation follows this progression:\n\n1. inspect the benchmark and registered raw datasets;\n2. generate data-processing code and a training config;\n3. fail fast on schema\u002Fconfig\u002Fruntime errors;\n4. fine-tune the target model;\n5. evaluate with the configured evaluation stack;\n6. feed validation feedback into the next iteration while keeping test results for final reporting.\n\nThe `--dataset` option selects the agent's dataset after registered resources are prepared; it does not necessarily avoid initial dataset preparation. Reuse a stable `FT_FILE_PATH` when intentionally caching assets.\n\n## Batch jobs and UI\n\nFor repeatable multi-task runs, copy the checked-in task template to a private file, copy the private `.env` into the job directory, and run the bundled job shell script only after inspecting its commands. The Streamlit fine-tune UI is a reporting\u002Finteraction surface, not proof that training or evaluation succeeded. Preserve training config, dataset schema, checkpoint path, merge\u002Fevaluation output, and validation metric.\n\n## Failure boundaries\n\n- dataset\u002Fdownload\u002Flicense failure → data\u002Fprovider block;\n- generated data schema failure → fix processing before training;\n- Docker\u002FCUDA\u002Fmodel-load failure → backend block;\n- training completes but evaluation cannot consume the checkpoint → artifact\u002Fevaluator failure;\n- provider quota or context failure → provider failure, not a model-quality result.\n\nRead [fine-tune-contract.md](references\u002Ffine-tune-contract.md) before a costly run and the parent troubleshooting guide for recovery.\n","# RD-Agent LLM fine-tuning\n\nUse this sub-skill for `rdagent llm_finetune`, FT-Agent datasets\u002Fbenchmarks, LLaMA-Factory training configuration, OpenCompass evaluation, batch jobs, and the fine-tune UI.\n\n## Prerequisites and cost gate\n\nA realistic run can download models and datasets, call an LLM provider, start Docker or conda training environments, and consume GPU hours. Before running, confirm Linux, a usable Docker\u002Fconda backend, NVIDIA\u002FCUDA capacity when required, Hugging Face access, provider credentials, and a stable `FT_FILE_PATH`. A missing prerequisite is a blocked run, not a reason to substitute a toy score.\n\nKeep the main settings explicit and secret-free in the shell or a private `.env`:\n\n```text\nBACKEND\nCHAT_MODEL\nOPENAI_API_KEY \u002F OPENAI_API_BASE\nEMBEDDING_MODEL\nFT_FILE_PATH\nFT_Coder_CoSTEER_env_type=docker|conda\nFT_TARGET_BENCHMARK\nFT_BENCHMARK_DESCRIPTION\nFT_BASE_MODEL\nFT_UPPER_DATA_SIZE_LIMIT\nFT_API_MAX_WORKERS\nFT_STRONG_MODELS \u002F FT_WEAK_MODELS\nHF_TOKEN (only when required)\n```\n\nUse [validate_ft_env.py](scripts\u002Fvalidate_ft_env.py) to print presence only; it never prints values.\n\n## Single-task workflow\n\nStart with help and a bounded configuration:\n\n```bash\nrdagent llm_finetune --help\nrdagent llm_finetune \\\n  --benchmark \u003Cbenchmark> \\\n  --benchmark-description \"\u003Ctask and output format>\" \\\n  --base-model \u003Chuggingface-model-id> \\\n  --loop-n 1 \\\n  --timeout \u003Cshort-budget>\n```\n\nThe checked-in implementation follows this progression:\n\n1. inspect the benchmark and registered raw datasets;\n2. generate data-processing code and a training config;\n3. fail fast on schema\u002Fconfig\u002Fruntime errors;\n4. fine-tune the target model;\n5. evaluate with the configured evaluation stack;\n6. feed validation feedback into the next iteration while keeping test results for final reporting.\n\nThe `--dataset` option selects the agent's dataset after registered resources are prepared; it does not necessarily avoid initial dataset preparation. Reuse a stable `FT_FILE_PATH` when intentionally caching assets.\n\n## Batch jobs and UI\n\nFor repeatable multi-task runs, copy the checked-in task template to a private file, copy the private `.env` into the job directory, and run the bundled job shell script only after inspecting its commands. The Streamlit fine-tune UI is a reporting\u002Finteraction surface, not proof that training or evaluation succeeded. Preserve training config, dataset schema, checkpoint path, merge\u002Fevaluation output, and validation metric.\n\n## Failure boundaries\n\n- dataset\u002Fdownload\u002Flicense failure → data\u002Fprovider block;\n- generated data schema failure → fix processing before training;\n- Docker\u002FCUDA\u002Fmodel-load failure → backend block;\n- training completes but evaluation cannot consume the checkpoint → artifact\u002Fevaluator failure;\n- provider quota or context failure → provider failure, not a model-quality result.\n\nRead [fine-tune-contract.md](references\u002Ffine-tune-contract.md) before a costly run and the parent troubleshooting guide for recovery.\n",{"name":5,"license":15,"metadata":24,"description":7,"disable-model-invocation":27},{"disco-role":25,"parent-skill":26},"operating","rd-agent",true,"main","d69096e239f2a89aabc2cd078c47664a9d3b1e11","05-技能与代理开发\u002Frd-agent\u002Fsub-skills\u002Fllm-finetune",3,"https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts","https:\u002F\u002Fgithub.com\u002Flza6","个人vx：Tf00798\r\n个人✈：@TFGY999","2026-10-08 14:29:22",{"source":8,"skillName":5,"commands":37,"prompt":47,"zipAvailable":27},[38,41,44],{"tool":39,"command":40},"npx","npx skills add lza6\u002FArmor-piercing-system-prompts --skill llm-finetune",{"tool":42,"command":43},"bunx","bunx skills add lza6\u002FArmor-piercing-system-prompts --skill llm-finetune",{"tool":45,"command":46},"pnpm","pnpm dlx skills add lza6\u002FArmor-piercing-system-prompts --skill llm-finetune","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F05-技能与代理开发\u002Frd-agent\u002Fsub-skills\u002Fllm-finetune\u002FSKILL.md",[49],{"path":50,"resourceType":51,"resourceTypeDesc":52,"sizeBytes":53,"sourceHtmlUrl":54},"references\u002Ffine-tune-contract.md","REFERENCE","参考文档",1274,"https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F05-技能与代理开发\u002Frd-agent\u002Fsub-skills\u002Fllm-finetune\u002Freferences\u002Ffine-tune-contract.md",[56,63,68,75,82,89],{"skillNo":57,"name":58,"nameZh":6,"description":59,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":60,"categories":61,"keywords":62,"publishedAt":19},"01M7FFG485KY83TD6HSC4YPCMC","obsidian-clipper-template-creator","Guide for creating templates for the Obsidian Web Clipper. Use when you want to create a new clipping template, understand available variables, or format clipped content.","https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F07-后端与架构\u002Fobsidian-clipper-template-creator\u002FSKILL.md",[],[],{"skillNo":64,"name":58,"nameZh":6,"description":59,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":65,"categories":66,"keywords":67,"publishedAt":19},"01M7FFG0T6AZ0STEJPSB3CX84K","https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F07-后端与架构\u002Fobsidian-clipper-template-creator-antigravity-awesome-skills-main\u002FSKILL.md",[],[],{"skillNo":69,"name":70,"nameZh":6,"description":71,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":72,"categories":73,"keywords":74,"publishedAt":19},"01M7FFFY0QEM14SKWPDYGMXM2P","python-api","Programmatic Object-Detection-Metrics API guidance for BoundingBox, BoundingBoxes, Evaluator, enums, coordinate conversion, AP\u002FmAP semantics, and source-style import diagnostics.","https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F07-后端与架构\u002Fobject-detection-metrics\u002Fsub-skills\u002Fpython-api\u002FSKILL.md",[],[],{"skillNo":76,"name":77,"nameZh":6,"description":78,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":79,"categories":80,"keywords":81,"publishedAt":19},"01M7FFFVZQZ2GJ0C0TGP7JHD20","file-evaluation","Folder-based PASCAL VOC AP\u002FmAP evaluation from ground-truth and detection text files.","https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F07-后端与架构\u002Fobject-detection-metrics\u002Fsub-skills\u002Ffile-evaluation\u002FSKILL.md",[],[],{"skillNo":83,"name":84,"nameZh":6,"description":85,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":86,"categories":87,"keywords":88,"publishedAt":19},"01M7FFFSFNYSE77SJ6PE1CK2DA","object-detection-metrics","Guides legacy Object-Detection-Metrics PASCAL VOC AP\u002FmAP evaluation from detection text folders and direct Python API classes.","https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F07-后端与架构\u002Fobject-detection-metrics\u002FSKILL.md",[],[],{"skillNo":90,"name":91,"nameZh":6,"description":92,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":93,"categories":94,"keywords":95,"publishedAt":19},"01M7FFFP0GCK1RVXBCPQ8APBMT","nx-workspace-patterns","Configure and optimize Nx monorepo workspaces. Use when setting up Nx, configuring project boundaries, optimizing build caching, or implementing affected commands.","https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F07-后端与架构\u002Fnx-workspace-patterns\u002FSKILL.md",[],[]]