act-so101-pen

name	act-so101-pen
description	S1 Vision-Language-Action policy. Capabilities: pick, place, pick_and_place, transfer on pen. ACT (Zhao et al., 2023) finetuned for "pass the pen" pick-and-place on a real SO-101 follower arm (Apache-2.0). ResNet-18 backbone, 4+1 enc/dec, latent VAE, chunk_size=100. Emits 6-DoF absolute joint-position chunks (in joint degrees) from two RGB views (front/overview + wrist). A smaller, faster, ONNX/TensorRT-friendly sibling of rskill-smolvla-so101-pen-bf16; the whole model exports to a single ONNX graph (OPENRAL_ACT_TRT=1). Discovery view of an OpenRAL rSkill — NOT directly runnable by an agent harness; it runs via rSkill.from_pretrained + the robot HAL.
metadata	{"openral_rskill":true,"schema_version":0.1,"rskill_id":"OpenRAL/rskill-act-so101-pen-fp32","manifest":"./rskill.yaml","role":"s1","kind":"vla","model_family":"act","embodiment_tags":["so101_follower"],"actions":["pick","place","pick_and_place","transfer"],"objects":["pen"],"scenes":["tabletop"],"sensors_required":["rgb:observation.images.camera1","rgb:observation.images.camera2"],"state_dim":6,"action_dim":6,"action_representation":"joint_positions","runtime":"pytorch","quantization":"fp32/pytorch","min_vram_gb":{"fp32":0.5},"chunk_size":100,"latency_budget":{"per_chunk_ms":100},"license_code":"Apache-2.0","license_weights":"apache-2.0","weights_uri":"hf://gabrycina/so101-passing-pen-policy","source_repo":"hf://gabrycina/so101-passing-pen-policy","paper_url":"https://arxiv.org/abs/2304.13705"}

act-so101-pen — rSkill discovery view

Generated view, not a hand-written skill. This SKILL.md is a discovery-only mirror of rskill.yaml, produced by tools/generate_rskill_skillmd.py. It lets tools that read the standard agent-skill format find and reason about this OpenRAL rSkill. The rskill.yaml manifest is the single source of truth (CLAUDE.md §1.3). Do not edit by hand — edit the manifest and regenerate.

What it is

An OpenRAL Vision-Language-Action policy (role: s1, kind: vla). ACT (Zhao et al., 2023) finetuned for "pass the pen" pick-and-place on a real SO-101 follower arm (Apache-2.0). ResNet-18 backbone, 4+1 enc/dec, latent VAE, chunk_size=100. Emits 6-DoF absolute joint-position chunks (in joint degrees) from two RGB views (front/overview + wrist). A smaller, faster, ONNX/TensorRT-friendly sibling of rskill-smolvla-so101-pen-bf16; the whole model exports to a single ONNX graph (OPENRAL_ACT_TRT=1).

Capabilities

• Verbs: pick · place · pick_and_place · transfer
• Objects: pen
• Scenes: tabletop
• Embodiments: so101_follower

Why this is discovery-only

An agent skill is natural-language instructions loaded into an LLM's context. An rSkill is an executable artifact: it carries a typed capability/embodiment contract, model weights, a runtime, and a license/provenance gate — none of which fit in freeform markdown. So an agent can use this view to select the right skill, but cannot execute it by loading this file. Execution always goes through the OpenRAL loader and the robot HAL.

License

• Code: Apache-2.0.
• Weights: apache-2.0 — permissive / commercial-use OK

How to actually run it (not via an agent harness)

from openral_rskill import rSkill

skill = rSkill.from_pretrained("OpenRAL/rskill-act-so101-pen-fp32")
# the loader validates embodiment / sensors / runtime / quantization against the target
# RobotDescription and enforces the weight-license gate before any weights load.

See rskill.yaml for the authoritative, validated manifest.