[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M78ZJ2WP6VWMACT34JTCPW0R":3,"skill-same:01M78ZJ2WP6VWMACT34JTCPW0R":70},{"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},"01M78ZJ2WP6VWMACT34JTCPW0R","prompt-optimization",null,"Operate ProTeGi text prompt optimization and Promptist text-to-image prompt rewriting workflows for LMOps.","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\u002Flmops\u002Fsub-skills\u002Fprompt-optimization\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: prompt-optimization\ndescription: \"Operate ProTeGi text prompt optimization and Promptist\n  text-to-image prompt rewriting workflows for LMOps.\"\ndisable-model-invocation: true\nmetadata:\n  disco-role: operating\nlicense: MIT\n---\n\n# Prompt Optimization\n\nUse this operating sub-skill when the user asks for LMOps prompt optimization, prompt rewriting, or prompt-optimization run planning.\n\n## Route by task shape\n\n- **ProTeGi text prompt optimization**: use for automatic optimization of zero-shot binary classification prompts with labeled train\u002Ftest data, markdown seed prompts, and an approved chat\u002Fcompletion provider budget.\n- **Promptist prompt rewriting**: use for text-to-image prompt rewriting with the pretrained Promptist prompter model, but plan offline first to avoid accidental model downloads.\n- **Promptist RL training**: use only for planning or explicitly approved heavy execution; it needs training prompt data, a supervised prompter checkpoint, Stable Diffusion\u002FCLIP\u002Faesthetic reward assets, Accelerate\u002FTRL, GPUs, and credentials.\n- **Retrieval-based prompt or demonstration selection**: route to `..\u002Fexample-retrieval\u002FSKILL.md`.\n- **Domain\u002Finstruction corpus conversion or model adaptation before prompt work**: route to `..\u002Fadaptation-and-training\u002FSKILL.md`.\n- **Broad LMOps project routing**: consult `..\u002F..\u002Freferences\u002Fproject-index.md` if the user names a different LMOps project.\n\n## Required first checks\n\n1. Classify the request as ProTeGi optimization, Promptist pretrained rewrite planning, Promptist RL training planning, or a route-out case.\n2. Confirm required inputs before any expensive or external action:\n   - ProTeGi: task name, task data directory, seed prompt markdown file(s), output log path, evaluator, scorer, provider credentials, overwrite consent, and budget\u002Fconcurrency limits.\n   - Promptist rewrite: plain text prompt(s), model id, tokenizer id, model cache or network consent, and CPU\u002FGPU expectation for any real model load.\n   - Promptist RL training: data directory, supervised prompter checkpoint, Stable Diffusion model access, TRL\u002FPPO config, checkpoint directory, distributed launch plan, GPU availability, Hugging Face access, and W&B\u002Flogging policy.\n3. Use the bundled safe planners before any source checkout run:\n   - `python scripts\u002Fprotegi_command_builder.py --help`\n   - `python scripts\u002Fpromptist_rewrite_skeleton.py --help`\n4. Read the focused runtime references:\n   - `references\u002Fprotegi-cli-and-api.md`\n   - `references\u002Fpromptist-workflows.md`\n   - `references\u002Ftroubleshooting.md`\n\n## Safe operating defaults\n\n- Do not run model downloads, provider calls, image generation, PPO training, Docker\u002Fcontainer startup, or multi-node launches unless the user explicitly asks and provides the needed environment and credentials.\n- Treat bundled scripts as planners and validators. They do not import LMOps source code, call model providers, load Promptist models, download weights, generate images, or train.\n- Promptist GPU\u002Fmodel-download workflows and multi-node RL training are documented but not creation-time verified by this sub-skill.\n- Treat ProTeGi output files as user data: the native program removes an existing output file before writing, so ask before overwriting.\n\n## Quick ProTeGi planning pattern\n\n1. Choose one of the supported native task names: `ethos`, `jailbreak`, `liar`, or `ar_sarcasm`.\n2. Stage task data in the required layout and one or more seed prompt markdown files with a `# Task` section and `{{ text }}` placeholder.\n3. Build a command without executing it:\n\n```bash\npython scripts\u002Fprotegi_command_builder.py \\\n  --task liar \\\n  --data-dir data\u002Fmy_binary_task \\\n  --prompts prompts\u002Fseed.md \\\n  --out runs\u002Fmy_binary_task.ucb.out \\\n  --evaluator ucb \\\n  --scorer 01 \\\n  --path-policy warn\n```\n\n4. Resolve any warnings, confirm credentials and overwrite behavior, then run the emitted command only in a prepared ProTeGi environment.\n\n## Quick Promptist planning pattern\n\n1. Start with the offline skeleton and keep model loading off:\n\n```bash\npython scripts\u002Fpromptist_rewrite_skeleton.py \\\n  --plain-text \"A rabbit is wearing a space suit\" \\\n  --model-id microsoft\u002FPromptist \\\n  --tokenizer-id gpt2 \\\n  --show-prompts\n```\n\n2. If the user approves a real rewrite run, confirm PyTorch\u002FTransformers availability and whether the Promptist model and tokenizer are already cached or may be downloaded.\n3. If the user asks for RL training, move to the training checklist in `references\u002Fpromptist-workflows.md` and do not treat it as a small smoke test.\n","# Prompt Optimization\n\nUse this operating sub-skill when the user asks for LMOps prompt optimization, prompt rewriting, or prompt-optimization run planning.\n\n## Route by task shape\n\n- **ProTeGi text prompt optimization**: use for automatic optimization of zero-shot binary classification prompts with labeled train\u002Ftest data, markdown seed prompts, and an approved chat\u002Fcompletion provider budget.\n- **Promptist prompt rewriting**: use for text-to-image prompt rewriting with the pretrained Promptist prompter model, but plan offline first to avoid accidental model downloads.\n- **Promptist RL training**: use only for planning or explicitly approved heavy execution; it needs training prompt data, a supervised prompter checkpoint, Stable Diffusion\u002FCLIP\u002Faesthetic reward assets, Accelerate\u002FTRL, GPUs, and credentials.\n- **Retrieval-based prompt or demonstration selection**: route to `..\u002Fexample-retrieval\u002FSKILL.md`.\n- **Domain\u002Finstruction corpus conversion or model adaptation before prompt work**: route to `..\u002Fadaptation-and-training\u002FSKILL.md`.\n- **Broad LMOps project routing**: consult `..\u002F..\u002Freferences\u002Fproject-index.md` if the user names a different LMOps project.\n\n## Required first checks\n\n1. Classify the request as ProTeGi optimization, Promptist pretrained rewrite planning, Promptist RL training planning, or a route-out case.\n2. Confirm required inputs before any expensive or external action:\n   - ProTeGi: task name, task data directory, seed prompt markdown file(s), output log path, evaluator, scorer, provider credentials, overwrite consent, and budget\u002Fconcurrency limits.\n   - Promptist rewrite: plain text prompt(s), model id, tokenizer id, model cache or network consent, and CPU\u002FGPU expectation for any real model load.\n   - Promptist RL training: data directory, supervised prompter checkpoint, Stable Diffusion model access, TRL\u002FPPO config, checkpoint directory, distributed launch plan, GPU availability, Hugging Face access, and W&B\u002Flogging policy.\n3. Use the bundled safe planners before any source checkout run:\n   - `python scripts\u002Fprotegi_command_builder.py --help`\n   - `python scripts\u002Fpromptist_rewrite_skeleton.py --help`\n4. Read the focused runtime references:\n   - `references\u002Fprotegi-cli-and-api.md`\n   - `references\u002Fpromptist-workflows.md`\n   - `references\u002Ftroubleshooting.md`\n\n## Safe operating defaults\n\n- Do not run model downloads, provider calls, image generation, PPO training, Docker\u002Fcontainer startup, or multi-node launches unless the user explicitly asks and provides the needed environment and credentials.\n- Treat bundled scripts as planners and validators. They do not import LMOps source code, call model providers, load Promptist models, download weights, generate images, or train.\n- Promptist GPU\u002Fmodel-download workflows and multi-node RL training are documented but not creation-time verified by this sub-skill.\n- Treat ProTeGi output files as user data: the native program removes an existing output file before writing, so ask before overwriting.\n\n## Quick ProTeGi planning pattern\n\n1. Choose one of the supported native task names: `ethos`, `jailbreak`, `liar`, or `ar_sarcasm`.\n2. Stage task data in the required layout and one or more seed prompt markdown files with a `# Task` section and `{{ text }}` placeholder.\n3. Build a command without executing it:\n\n```bash\npython scripts\u002Fprotegi_command_builder.py \\\n  --task liar \\\n  --data-dir data\u002Fmy_binary_task \\\n  --prompts prompts\u002Fseed.md \\\n  --out runs\u002Fmy_binary_task.ucb.out \\\n  --evaluator ucb \\\n  --scorer 01 \\\n  --path-policy warn\n```\n\n4. Resolve any warnings, confirm credentials and overwrite behavior, then run the emitted command only in a prepared ProTeGi environment.\n\n## Quick Promptist planning pattern\n\n1. Start with the offline skeleton and keep model loading off:\n\n```bash\npython scripts\u002Fpromptist_rewrite_skeleton.py \\\n  --plain-text \"A rabbit is wearing a space suit\" \\\n  --model-id microsoft\u002FPromptist \\\n  --tokenizer-id gpt2 \\\n  --show-prompts\n```\n\n2. If the user approves a real rewrite run, confirm PyTorch\u002FTransformers availability and whether the Promptist model and tokenizer are already cached or may be downloaded.\n3. If the user asks for RL training, move to the training checklist in `references\u002Fpromptist-workflows.md` and do not treat it as a small smoke test.\n",{"name":5,"license":14,"metadata":23,"description":7,"disable-model-invocation":25},{"disco-role":24},"operating",true,"main","ac3fe1afa80fb9a09775ecfb2b6cc3ba850a2db6","skills\u002Frepositories\u002Frepo-skills\u002Flmops\u002Fsub-skills\u002Fprompt-optimization",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 prompt-optimization",{"tool":39,"command":40},"bunx","bunx skills add VectorSpaceLab\u002FAREX-Skill --skill prompt-optimization",{"tool":42,"command":43},"pnpm","pnpm dlx skills add VectorSpaceLab\u002FAREX-Skill --skill prompt-optimization","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Flmops\u002Fsub-skills\u002Fprompt-optimization\u002FSKILL.md",[46,52,56,60,66],{"path":47,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":50,"sourceHtmlUrl":51},"references\u002Fpromptist-workflows.md","REFERENCE","参考文档",5410,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Flmops\u002Fsub-skills\u002Fprompt-optimization\u002Freferences\u002Fpromptist-workflows.md",{"path":53,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":54,"sourceHtmlUrl":55},"references\u002Fprotegi-cli-and-api.md",9119,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Flmops\u002Fsub-skills\u002Fprompt-optimization\u002Freferences\u002Fprotegi-cli-and-api.md",{"path":57,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":58,"sourceHtmlUrl":59},"references\u002Ftroubleshooting.md",6086,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Flmops\u002Fsub-skills\u002Fprompt-optimization\u002Freferences\u002Ftroubleshooting.md",{"path":61,"resourceType":62,"resourceTypeDesc":63,"sizeBytes":64,"sourceHtmlUrl":65},"scripts\u002Fpromptist_rewrite_skeleton.py","SCRIPT","脚本",8316,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Flmops\u002Fsub-skills\u002Fprompt-optimization\u002Fscripts\u002Fpromptist_rewrite_skeleton.py",{"path":67,"resourceType":62,"resourceTypeDesc":63,"sizeBytes":68,"sourceHtmlUrl":69},"scripts\u002Fprotegi_command_builder.py",13038,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Flmops\u002Fsub-skills\u002Fprompt-optimization\u002Fscripts\u002Fprotegi_command_builder.py",[71,79,86,93,100,107],{"skillNo":72,"name":73,"nameZh":6,"description":74,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":75,"licenseFileUrl":6,"sourceHtmlUrl":76,"categories":77,"keywords":78,"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":80,"name":81,"nameZh":6,"description":82,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":75,"licenseFileUrl":6,"sourceHtmlUrl":83,"categories":84,"keywords":85,"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":87,"name":88,"nameZh":6,"description":89,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":75,"licenseFileUrl":6,"sourceHtmlUrl":90,"categories":91,"keywords":92,"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":94,"name":95,"nameZh":6,"description":96,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":75,"licenseFileUrl":6,"sourceHtmlUrl":97,"categories":98,"keywords":99,"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":101,"name":102,"nameZh":6,"description":103,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":75,"licenseFileUrl":6,"sourceHtmlUrl":104,"categories":105,"keywords":106,"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":108,"name":109,"nameZh":6,"description":110,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":75,"licenseFileUrl":6,"sourceHtmlUrl":111,"categories":112,"keywords":113,"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",[],[]]