[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M79132XJT8WNXVR8CFY9C6RP":3,"skill-same:01M79132XJT8WNXVR8CFY9C6RP":71},{"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},"01M79132XJT8WNXVR8CFY9C6RP","agents",null,"Implement and debug NAVSIM rule-based, privileged, learned, TransFuser, and latent TransFuser agents while preserving sensor, trajectory, feature-builder, and checkpoint contracts.","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\u002Fnavsim\u002Fsub-skills\u002Fagents\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: agents\ndescription: \"Implement and debug NAVSIM rule-based, privileged, learned,\n  TransFuser, and latent TransFuser agents while preserving sensor, trajectory,\n  feature-builder, and checkpoint contracts.\"\ndisable-model-invocation: true\nmetadata:\n  disco-role: operating\nlicense: Apache 2.0\n---\n\n# NAVSIM agents\n\nUse this route when an agent must be selected, implemented, trained, loaded, or\ntroubleshot. It covers the `AbstractAgent` interface, rule-based and privileged\nbaselines, ego-status and TransFuser\u002FLTF learned agents, sensor history, local\ntrajectory sampling, feature\u002Ftarget builders, and checkpoint loading.\n\n## Route by intent\n\n- **Interface or API design:** read [API reference](references\u002Fapi-reference.md).\n- **Sensor selection or output shape:** read\n  [trajectory and sensors](references\u002Ftrajectory-and-sensors.md).\n- **Baseline choice or learned-agent extension:** read\n  [baselines](references\u002Fbaselines.md).\n- **An error or failed run:** read [troubleshooting](references\u002Ftroubleshooting.md)\n  and run the safe validator before touching data or launching a workload.\n\n## Required operating rules\n\n1. Subclass `AbstractAgent`, expose a stable `name()`, initialize once per\n   worker, return a deliberate `SensorConfig`, and return local rear-axle BEV\n   `(x, y, heading)` poses in a `Trajectory`.\n2. Treat the trajectory sampling object as part of the model contract. The\n   repository default requests a 4-second horizon sampled every 0.5 seconds;\n   evaluation uses a 4-second, 0.1-second proposal. Do not silently change one\n   to satisfy the other.\n3. A submission receives `AgentInput`, not an annotated `Scene`. Keep\n   `requires_scene=False` for submission agents; the human\u002Fprivileged baseline\n   is useful for analysis but is prohibited in submission creation.\n4. Choose sensors explicitly. TransFuser uses the current front camera trio and\n   current LiDAR by default; latent TransFuser replaces the LiDAR branch with a\n   learned latent tensor and must not request or build a LiDAR feature.\n5. Learned agents must keep feature-builder keys, target-builder outputs,\n   `forward()` output (including `\"trajectory\"` with shape `[B, T, 3]`), loss,\n   optimizer, and checkpoint key namespaces mutually compatible.\n\n## Safe verification\n\nRun the bundled [agent contract validator](scripts\u002Fvalidate_agent_contract.py):\n\n```bash\npython scripts\u002Fvalidate_agent_contract.py\npython scripts\u002Fvalidate_agent_contract.py --help\n```\n\nThe validator uses synthetic arrays and sensor configurations only. It does not\nrequire a dataset, maps, checkpoints, downloads, submission credentials, or a\nbenchmark. Add `--module path\u002Fto\u002Fyour_agent.py` to inspect importable agent\nclasses without instantiating or running them; use `--strict` when missing\noptional NAVSIM dependencies should fail the check.\n\nKeep training, scoring, submission generation, and dataset access on their\nrespective workflow routes. This sub-skill explains how an agent participates\nin those workflows but does not launch them by default.\n","# NAVSIM agents\n\nUse this route when an agent must be selected, implemented, trained, loaded, or\ntroubleshot. It covers the `AbstractAgent` interface, rule-based and privileged\nbaselines, ego-status and TransFuser\u002FLTF learned agents, sensor history, local\ntrajectory sampling, feature\u002Ftarget builders, and checkpoint loading.\n\n## Route by intent\n\n- **Interface or API design:** read [API reference](references\u002Fapi-reference.md).\n- **Sensor selection or output shape:** read\n  [trajectory and sensors](references\u002Ftrajectory-and-sensors.md).\n- **Baseline choice or learned-agent extension:** read\n  [baselines](references\u002Fbaselines.md).\n- **An error or failed run:** read [troubleshooting](references\u002Ftroubleshooting.md)\n  and run the safe validator before touching data or launching a workload.\n\n## Required operating rules\n\n1. Subclass `AbstractAgent`, expose a stable `name()`, initialize once per\n   worker, return a deliberate `SensorConfig`, and return local rear-axle BEV\n   `(x, y, heading)` poses in a `Trajectory`.\n2. Treat the trajectory sampling object as part of the model contract. The\n   repository default requests a 4-second horizon sampled every 0.5 seconds;\n   evaluation uses a 4-second, 0.1-second proposal. Do not silently change one\n   to satisfy the other.\n3. A submission receives `AgentInput`, not an annotated `Scene`. Keep\n   `requires_scene=False` for submission agents; the human\u002Fprivileged baseline\n   is useful for analysis but is prohibited in submission creation.\n4. Choose sensors explicitly. TransFuser uses the current front camera trio and\n   current LiDAR by default; latent TransFuser replaces the LiDAR branch with a\n   learned latent tensor and must not request or build a LiDAR feature.\n5. Learned agents must keep feature-builder keys, target-builder outputs,\n   `forward()` output (including `\"trajectory\"` with shape `[B, T, 3]`), loss,\n   optimizer, and checkpoint key namespaces mutually compatible.\n\n## Safe verification\n\nRun the bundled [agent contract validator](scripts\u002Fvalidate_agent_contract.py):\n\n```bash\npython scripts\u002Fvalidate_agent_contract.py\npython scripts\u002Fvalidate_agent_contract.py --help\n```\n\nThe validator uses synthetic arrays and sensor configurations only. It does not\nrequire a dataset, maps, checkpoints, downloads, submission credentials, or a\nbenchmark. Add `--module path\u002Fto\u002Fyour_agent.py` to inspect importable agent\nclasses without instantiating or running them; use `--strict` when missing\noptional NAVSIM dependencies should fail the check.\n\nKeep training, scoring, submission generation, and dataset access on their\nrespective workflow routes. This sub-skill explains how an agent participates\nin those workflows but does not launch them by default.\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\u002Fnavsim\u002Fsub-skills\u002Fagents",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 agents",{"tool":40,"command":41},"bunx","bunx skills add VectorSpaceLab\u002FAREX-Skill --skill agents",{"tool":43,"command":44},"pnpm","pnpm dlx skills add VectorSpaceLab\u002FAREX-Skill --skill agents","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fnavsim\u002Fsub-skills\u002Fagents\u002FSKILL.md",[47,53,57,61,65],{"path":48,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":51,"sourceHtmlUrl":52},"references\u002Fapi-reference.md","REFERENCE","参考文档",5365,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fnavsim\u002Fsub-skills\u002Fagents\u002Freferences\u002Fapi-reference.md",{"path":54,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":55,"sourceHtmlUrl":56},"references\u002Fbaselines.md",4672,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fnavsim\u002Fsub-skills\u002Fagents\u002Freferences\u002Fbaselines.md",{"path":58,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":59,"sourceHtmlUrl":60},"references\u002Ftrajectory-and-sensors.md",4794,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fnavsim\u002Fsub-skills\u002Fagents\u002Freferences\u002Ftrajectory-and-sensors.md",{"path":62,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":63,"sourceHtmlUrl":64},"references\u002Ftroubleshooting.md",7947,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fnavsim\u002Fsub-skills\u002Fagents\u002Freferences\u002Ftroubleshooting.md",{"path":66,"resourceType":67,"resourceTypeDesc":68,"sizeBytes":69,"sourceHtmlUrl":70},"scripts\u002Fvalidate_agent_contract.py","SCRIPT","脚本",9365,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fnavsim\u002Fsub-skills\u002Fagents\u002Fscripts\u002Fvalidate_agent_contract.py",[72,79,86,93,100,107],{"skillNo":73,"name":74,"nameZh":6,"description":75,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"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.","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":14,"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":14,"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":14,"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":14,"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":14,"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",[],[]]