[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M794C44KPARDKVBAZX7PTKP7":3,"skill-same:01M794C44KPARDKVBAZX7PTKP7":68},{"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":28,"commitSha":29,"skillDirectory":30,"forksCount":31,"repositoryHtmlUrl":32,"repositoryHomepageUrl":6,"ownerHtmlUrl":33,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":34,"install":35,"resources":47},"01M794C44KPARDKVBAZX7PTKP7","evaluation",null,"Use rLLM evaluation APIs and CLI for AgentFlow\u002Fevaluator authoring, benchmark execution, harness selection, sandboxed eval, saved episodes, and result interpretation.","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\u002Frllm\u002Fsub-skills\u002Fevaluation\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: evaluation\nsummary: \"Run rLLM evaluations and author AgentFlow\u002FEvaluator protocols.\"\ndescription: \"Use rLLM evaluation APIs and CLI for AgentFlow\u002Fevaluator\n  authoring, benchmark execution, harness selection, sandboxed eval, saved\n  episodes, and result interpretation.\"\ndisable-model-invocation: true\nmetadata:\n  disco-role: operating\nlicense: Apache 2.0\n---\n\n# rLLM Evaluation\n\nUse this sub-skill when the task involves `rllm eval`, `@rllm.rollout`, `@rllm.evaluator`, `AgentFlow`, `Evaluator`, built-in\u002Fcustom harnesses, benchmark scoring, pass@k, saved episodes, or debugging an agent's interaction traces.\n\n## Start Here\n\n1. Read `references\u002Fagentflow-and-evaluator-api.md` before writing custom flows or evaluators. It captures current import locations and return-type contracts.\n2. Read `references\u002Fevaluation-workflows.md` for CLI recipes, provider\u002Fmodel selection, pass@k, sandboxed tasks, saved episode files, and programmatic `run_dataset` usage.\n3. Read `references\u002Ftroubleshooting.md` for common protocol, verifier, sandbox, provider, and result-schema failures.\n4. For data layout or dataset registry work, switch to `..\u002Fdatasets\u002FSKILL.md`. For provider\u002Fmodel setup, switch to `..\u002Fcli-ops\u002FSKILL.md`. For training rollouts, switch to `..\u002Ftraining\u002FSKILL.md`.\n\n## Core Decisions\n\n- **Is the agent a simple Python function?** Use `@rllm.rollout` and return an `Episode`, `Trajectory`, or `None`. A `None` return is valid when gateway traces will fill steps later.\n- **Does the evaluator score host-side or in a sandbox?** Host-side evaluators can be fixed with `--evaluator`; per-task\u002Fverifier metadata is used when task config declares sandbox or script\u002Fmodule verifiers.\n- **Does the task need a sandbox?** rLLM joins the flow's `needs_env`, the evaluator\u002Fverifier kind, and task metadata\u002Fenvironment directories. Do not provision sandboxes for tasks with no consumer.\n- **Is this eval-only or training data generation?** Eval-only paths can save episodes\u002Fresults; training paths must preserve token IDs\u002Flogprobs and route through the training sub-skill.\n\n## Safe Checks\n\nRun the bundled protocol smoke test to verify decorators and evaluator coercion without a model call:\n\n```bash\npython scripts\u002Fagentflow_eval_smoke.py\n```\n\nIt constructs a tiny `Task`, `AgentConfig`, `Trajectory`, and evaluator result; it does not call a provider or modify user state.\n\n## Avoid\n\n- Do not copy stale imports from old docs if they reference missing `rllm.sdk` modules.\n- Do not hard-code provider calls in evaluator tests; keep provider selection at the CLI\u002Fconfig boundary.\n- Do not use a host-side evaluator for tasks whose `dataset.toml`\u002F`task.toml` expects sandbox-side tests unless you intentionally override that behavior.\n","# rLLM Evaluation\n\nUse this sub-skill when the task involves `rllm eval`, `@rllm.rollout`, `@rllm.evaluator`, `AgentFlow`, `Evaluator`, built-in\u002Fcustom harnesses, benchmark scoring, pass@k, saved episodes, or debugging an agent's interaction traces.\n\n## Start Here\n\n1. Read `references\u002Fagentflow-and-evaluator-api.md` before writing custom flows or evaluators. It captures current import locations and return-type contracts.\n2. Read `references\u002Fevaluation-workflows.md` for CLI recipes, provider\u002Fmodel selection, pass@k, sandboxed tasks, saved episode files, and programmatic `run_dataset` usage.\n3. Read `references\u002Ftroubleshooting.md` for common protocol, verifier, sandbox, provider, and result-schema failures.\n4. For data layout or dataset registry work, switch to `..\u002Fdatasets\u002FSKILL.md`. For provider\u002Fmodel setup, switch to `..\u002Fcli-ops\u002FSKILL.md`. For training rollouts, switch to `..\u002Ftraining\u002FSKILL.md`.\n\n## Core Decisions\n\n- **Is the agent a simple Python function?** Use `@rllm.rollout` and return an `Episode`, `Trajectory`, or `None`. A `None` return is valid when gateway traces will fill steps later.\n- **Does the evaluator score host-side or in a sandbox?** Host-side evaluators can be fixed with `--evaluator`; per-task\u002Fverifier metadata is used when task config declares sandbox or script\u002Fmodule verifiers.\n- **Does the task need a sandbox?** rLLM joins the flow's `needs_env`, the evaluator\u002Fverifier kind, and task metadata\u002Fenvironment directories. Do not provision sandboxes for tasks with no consumer.\n- **Is this eval-only or training data generation?** Eval-only paths can save episodes\u002Fresults; training paths must preserve token IDs\u002Flogprobs and route through the training sub-skill.\n\n## Safe Checks\n\nRun the bundled protocol smoke test to verify decorators and evaluator coercion without a model call:\n\n```bash\npython scripts\u002Fagentflow_eval_smoke.py\n```\n\nIt constructs a tiny `Task`, `AgentConfig`, `Trajectory`, and evaluator result; it does not call a provider or modify user state.\n\n## Avoid\n\n- Do not copy stale imports from old docs if they reference missing `rllm.sdk` modules.\n- Do not hard-code provider calls in evaluator tests; keep provider selection at the CLI\u002Fconfig boundary.\n- Do not use a host-side evaluator for tasks whose `dataset.toml`\u002F`task.toml` expects sandbox-side tests unless you intentionally override that behavior.\n",{"name":5,"license":23,"summary":24,"metadata":25,"description":7,"disable-model-invocation":27},"Apache 2.0","Run rLLM evaluations and author AgentFlow\u002FEvaluator protocols.",{"disco-role":26},"operating",true,"main","ac3fe1afa80fb9a09775ecfb2b6cc3ba850a2db6","skills\u002Frepositories\u002Frepo-skills\u002Frllm\u002Fsub-skills\u002Fevaluation",27,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab","2026-09-03 10:25:21",{"source":8,"skillName":5,"commands":36,"prompt":46,"zipAvailable":27},[37,40,43],{"tool":38,"command":39},"npx","npx skills add VectorSpaceLab\u002FAREX-Skill --skill evaluation",{"tool":41,"command":42},"bunx","bunx skills add VectorSpaceLab\u002FAREX-Skill --skill evaluation",{"tool":44,"command":45},"pnpm","pnpm dlx skills add VectorSpaceLab\u002FAREX-Skill --skill evaluation","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Frllm\u002Fsub-skills\u002Fevaluation\u002FSKILL.md",[48,54,58,62],{"path":49,"resourceType":50,"resourceTypeDesc":51,"sizeBytes":52,"sourceHtmlUrl":53},"references\u002Fagentflow-and-evaluator-api.md","REFERENCE","参考文档",2469,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Frllm\u002Fsub-skills\u002Fevaluation\u002Freferences\u002Fagentflow-and-evaluator-api.md",{"path":55,"resourceType":50,"resourceTypeDesc":51,"sizeBytes":56,"sourceHtmlUrl":57},"references\u002Fevaluation-workflows.md",2747,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Frllm\u002Fsub-skills\u002Fevaluation\u002Freferences\u002Fevaluation-workflows.md",{"path":59,"resourceType":50,"resourceTypeDesc":51,"sizeBytes":60,"sourceHtmlUrl":61},"references\u002Ftroubleshooting.md",1874,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Frllm\u002Fsub-skills\u002Fevaluation\u002Freferences\u002Ftroubleshooting.md",{"path":63,"resourceType":64,"resourceTypeDesc":65,"sizeBytes":66,"sourceHtmlUrl":67},"scripts\u002Fagentflow_eval_smoke.py","SCRIPT","脚本",2045,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Frllm\u002Fsub-skills\u002Fevaluation\u002Fscripts\u002Fagentflow_eval_smoke.py",[69,76,83,90,97,104],{"skillNo":70,"name":71,"nameZh":6,"description":72,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":73,"categories":74,"keywords":75,"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":77,"name":78,"nameZh":6,"description":79,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":80,"categories":81,"keywords":82,"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":84,"name":85,"nameZh":6,"description":86,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":87,"categories":88,"keywords":89,"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":91,"name":92,"nameZh":6,"description":93,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":94,"categories":95,"keywords":96,"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":98,"name":99,"nameZh":6,"description":100,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":101,"categories":102,"keywords":103,"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":105,"name":106,"nameZh":6,"description":107,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":108,"categories":109,"keywords":110,"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",[],[]]