[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M794AXMY4PGN99MHVSSW9XES":3,"skill-same:01M794AXMY4PGN99MHVSSW9XES":66},{"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},"01M794AXMY4PGN99MHVSSW9XES","mpc-control",null,"Operate the CPU-side quadruped convex MPC controller, its robot\u002Fstate\u002Fcommand interfaces, gait and FSM runners, policy bridge, and compiled solver extension.","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\u002Frl-mpc-locomotion\u002Fsub-skills\u002Fmpc-control\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: mpc-control\ndescription: \"Operate the CPU-side quadruped convex MPC controller, its\n  robot\u002Fstate\u002Fcommand interfaces, gait and FSM runners, policy bridge, and\n  compiled solver extension.\"\ndisable-model-invocation: true\nmetadata:\n  disco-role: operating\nlicense: MIT\n---\n\n# MPC control\n\nUse this skill for low-level, CPU-side control questions: selecting a supported robot or\nrunner, converting state and commands into 12 joint torques, configuring gaits\u002FFSM\ntransitions, checking the `mpc_osqp` binding, or diagnosing controller data-shape\nfailures. Keep Isaac Gym lifecycle and Hydra\u002FRSL-RL training in the linked neighboring\nskills rather than reproducing them here.\n\n## Route first\n\n- Read [the API contract](references\u002Fapi-reference.md) before constructing records or\n  calling a runner. It is the source of truth for enum values, array shapes, frame\n  conventions, and policy boundaries.\n- Follow [the workflow](references\u002Fworkflows.md) for install\u002Fbuild order, CPU smoke\n  checks, gamepad-free operation, and the state-to-torque loop.\n- Use [troubleshooting](references\u002Ftroubleshooting.md) for extension import errors,\n  malformed simulator records, unsafe FSM transitions, and policy\u002Fcheckpoint failures.\n- For package installation, compiler\u002Fdependency ordering, or backend diagnostics, route\n  to [setup-and-diagnostics](..\u002Fsetup-and-diagnostics\u002FSKILL.md).\n- For simulator assets, DOF effort mode, rigid-body records, terrains, viewer loops, or\n  Isaac Gym availability, route to [isaac-gym-simulation](..\u002Fisaac-gym-simulation\u002FSKILL.md).\n  For policy training\u002Fconfiguration, route to [rl-training](..\u002Frl-training\u002FSKILL.md).\n\n## Operating contract\n\n1. Select one of `RobotType.ALIENGO`, `RobotType.A1`, or `RobotType.GO1`, and construct\n   `Quadruped(robot_type)`. Do not invent a fourth robot model; Mini Cheetah is not an\n   enabled public enum. Preserve the model's four-leg, three-joint-per-leg ordering.\n2. Select `ControllerType.FSM`, `ControllerType.MIN`, or `ControllerType.POLICY`.\n   `RobotRunnerFSM` adds state transitions, `RobotRunnerMin` calls convex MPC directly,\n   and `RobotRunnerPolicy` computes learned MPC weights before entering the FSM.\n3. Call `runner.init(robot_type)` once, then call `runner.run(dof_states, body_states,\n   commands)` once per control step. Reset between episodes with the runner's `reset()`.\n   The result is a NumPy `float32` vector of exactly 12 torques, ordered as four\n   consecutive three-joint leg blocks.\n4. Keep the input representation consistent. Normal simulator mode uses structured\n   DOF\u002Fbody records; `Parameters.bridge_MPC_to_RL` changes the low-level MPC path to\n   numeric arrays, but the policy runner still expects the structured DOF fields.\n5. Commands begin with `[x_velocity, y_velocity, yaw_rate]`. Optional MPC weights must\n   be nonnegative and have the exact documented length. A zero command is safe input,\n   not an implicit request to enter locomotion.\n6. Never claim a simulation or RL run is verified on an installation without Isaac Gym.\n   CPU imports, model construction, gait arithmetic, and the solver binding are\n   separately checkable; RL\u002Fsimulation remains a required-backend block when that\n   closed-source package cannot be imported.\n\n## Controller boundaries\n\n- `DesiredStateCommand` stores velocity\u002Fyaw commands and a 13-element MPC weight\n  vector. `ConvexMPCLocomotion` chooses the gait, estimates contact\u002Fground state,\n  solves for stance forces, and fills swing\u002Fstance leg commands.\n- `LegController` combines feed-forward force\u002Ftorque, Cartesian PD, and joint PD through\n  each leg Jacobian. `zeroCommand()` must precede each controller call so stale commands\n  cannot survive a failed branch; `updateCommand()` is the torque boundary.\n- FSM mode starts from `Parameters.control_mode`. In normal operation, recovery stand,\n  passive, and locomotion transition through `ControlFSM`; safety checks can redirect\n  unsafe locomotion to recovery. Test mode runs the current state without transition\n  checks.\n- The policy bridge is optional and heavyweight: it imports the RL\u002Fpolicy stack, needs a\n  valid checkpoint and CUDA policy runtime, produces 12 scaled nonnegative weights, and\n  appends the required thirteenth gravity-placeholder weight before MPC. It is not a\n  substitute for the convex-MPC smoke test.\n\n## Verification gate\n\nRun the bundled `scripts\u002Fcheck_mpc_api.py` as a read-only API\u002Fextension probe. Treat an\nextension import failure as a build\u002Fruntime diagnosis, not as evidence that the control\nmath is wrong. The prepared handoff passed Python 3.8, PyTorch 1.10\u002FCUDA 11.3 on an\nA100 allocation, package imports, `rsl_rl`, and `mpc_osqp`; Isaac Gym was unavailable,\nso RL\u002Fsimulation remains an explicit required-backend block. Do not run interactive\nnative demos or simulator tests as part of a CPU smoke check; they are deferred until\nthe required Isaac Gym backend is installed and verified. Keep review reports and test\nfixtures outside this runtime subtree. Preserve this separation when handing the\ncontroller to another application: API success is not a claim of gait stability, robot\nsafety, or sim-to-real readiness.\n","# MPC control\n\nUse this skill for low-level, CPU-side control questions: selecting a supported robot or\nrunner, converting state and commands into 12 joint torques, configuring gaits\u002FFSM\ntransitions, checking the `mpc_osqp` binding, or diagnosing controller data-shape\nfailures. Keep Isaac Gym lifecycle and Hydra\u002FRSL-RL training in the linked neighboring\nskills rather than reproducing them here.\n\n## Route first\n\n- Read [the API contract](references\u002Fapi-reference.md) before constructing records or\n  calling a runner. It is the source of truth for enum values, array shapes, frame\n  conventions, and policy boundaries.\n- Follow [the workflow](references\u002Fworkflows.md) for install\u002Fbuild order, CPU smoke\n  checks, gamepad-free operation, and the state-to-torque loop.\n- Use [troubleshooting](references\u002Ftroubleshooting.md) for extension import errors,\n  malformed simulator records, unsafe FSM transitions, and policy\u002Fcheckpoint failures.\n- For package installation, compiler\u002Fdependency ordering, or backend diagnostics, route\n  to [setup-and-diagnostics](..\u002Fsetup-and-diagnostics\u002FSKILL.md).\n- For simulator assets, DOF effort mode, rigid-body records, terrains, viewer loops, or\n  Isaac Gym availability, route to [isaac-gym-simulation](..\u002Fisaac-gym-simulation\u002FSKILL.md).\n  For policy training\u002Fconfiguration, route to [rl-training](..\u002Frl-training\u002FSKILL.md).\n\n## Operating contract\n\n1. Select one of `RobotType.ALIENGO`, `RobotType.A1`, or `RobotType.GO1`, and construct\n   `Quadruped(robot_type)`. Do not invent a fourth robot model; Mini Cheetah is not an\n   enabled public enum. Preserve the model's four-leg, three-joint-per-leg ordering.\n2. Select `ControllerType.FSM`, `ControllerType.MIN`, or `ControllerType.POLICY`.\n   `RobotRunnerFSM` adds state transitions, `RobotRunnerMin` calls convex MPC directly,\n   and `RobotRunnerPolicy` computes learned MPC weights before entering the FSM.\n3. Call `runner.init(robot_type)` once, then call `runner.run(dof_states, body_states,\n   commands)` once per control step. Reset between episodes with the runner's `reset()`.\n   The result is a NumPy `float32` vector of exactly 12 torques, ordered as four\n   consecutive three-joint leg blocks.\n4. Keep the input representation consistent. Normal simulator mode uses structured\n   DOF\u002Fbody records; `Parameters.bridge_MPC_to_RL` changes the low-level MPC path to\n   numeric arrays, but the policy runner still expects the structured DOF fields.\n5. Commands begin with `[x_velocity, y_velocity, yaw_rate]`. Optional MPC weights must\n   be nonnegative and have the exact documented length. A zero command is safe input,\n   not an implicit request to enter locomotion.\n6. Never claim a simulation or RL run is verified on an installation without Isaac Gym.\n   CPU imports, model construction, gait arithmetic, and the solver binding are\n   separately checkable; RL\u002Fsimulation remains a required-backend block when that\n   closed-source package cannot be imported.\n\n## Controller boundaries\n\n- `DesiredStateCommand` stores velocity\u002Fyaw commands and a 13-element MPC weight\n  vector. `ConvexMPCLocomotion` chooses the gait, estimates contact\u002Fground state,\n  solves for stance forces, and fills swing\u002Fstance leg commands.\n- `LegController` combines feed-forward force\u002Ftorque, Cartesian PD, and joint PD through\n  each leg Jacobian. `zeroCommand()` must precede each controller call so stale commands\n  cannot survive a failed branch; `updateCommand()` is the torque boundary.\n- FSM mode starts from `Parameters.control_mode`. In normal operation, recovery stand,\n  passive, and locomotion transition through `ControlFSM`; safety checks can redirect\n  unsafe locomotion to recovery. Test mode runs the current state without transition\n  checks.\n- The policy bridge is optional and heavyweight: it imports the RL\u002Fpolicy stack, needs a\n  valid checkpoint and CUDA policy runtime, produces 12 scaled nonnegative weights, and\n  appends the required thirteenth gravity-placeholder weight before MPC. It is not a\n  substitute for the convex-MPC smoke test.\n\n## Verification gate\n\nRun the bundled `scripts\u002Fcheck_mpc_api.py` as a read-only API\u002Fextension probe. Treat an\nextension import failure as a build\u002Fruntime diagnosis, not as evidence that the control\nmath is wrong. The prepared handoff passed Python 3.8, PyTorch 1.10\u002FCUDA 11.3 on an\nA100 allocation, package imports, `rsl_rl`, and `mpc_osqp`; Isaac Gym was unavailable,\nso RL\u002Fsimulation remains an explicit required-backend block. Do not run interactive\nnative demos or simulator tests as part of a CPU smoke check; they are deferred until\nthe required Isaac Gym backend is installed and verified. Keep review reports and test\nfixtures outside this runtime subtree. Preserve this separation when handing the\ncontroller to another application: API success is not a claim of gait stability, robot\nsafety, or sim-to-real readiness.\n",{"name":5,"license":14,"metadata":23,"description":7,"disable-model-invocation":25},{"disco-role":24},"operating",true,"main","ac3fe1afa80fb9a09775ecfb2b6cc3ba850a2db6","skills\u002Frepositories\u002Frepo-skills\u002Frl-mpc-locomotion\u002Fsub-skills\u002Fmpc-control",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 mpc-control",{"tool":39,"command":40},"bunx","bunx skills add VectorSpaceLab\u002FAREX-Skill --skill mpc-control",{"tool":42,"command":43},"pnpm","pnpm dlx skills add VectorSpaceLab\u002FAREX-Skill --skill mpc-control","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Frl-mpc-locomotion\u002Fsub-skills\u002Fmpc-control\u002FSKILL.md",[46,52,56,60],{"path":47,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":50,"sourceHtmlUrl":51},"references\u002Fapi-reference.md","REFERENCE","参考文档",8956,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Frl-mpc-locomotion\u002Fsub-skills\u002Fmpc-control\u002Freferences\u002Fapi-reference.md",{"path":53,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":54,"sourceHtmlUrl":55},"references\u002Ftroubleshooting.md",6710,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Frl-mpc-locomotion\u002Fsub-skills\u002Fmpc-control\u002Freferences\u002Ftroubleshooting.md",{"path":57,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":58,"sourceHtmlUrl":59},"references\u002Fworkflows.md",5782,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Frl-mpc-locomotion\u002Fsub-skills\u002Fmpc-control\u002Freferences\u002Fworkflows.md",{"path":61,"resourceType":62,"resourceTypeDesc":63,"sizeBytes":64,"sourceHtmlUrl":65},"scripts\u002Fcheck_mpc_api.py","SCRIPT","脚本",4499,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Frl-mpc-locomotion\u002Fsub-skills\u002Fmpc-control\u002Fscripts\u002Fcheck_mpc_api.py",[67,75,82,89,96,103],{"skillNo":68,"name":69,"nameZh":6,"description":70,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":71,"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.","Apache-2.0","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":71,"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":71,"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":71,"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":71,"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":71,"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",[],[]]