Updated 1d ago
by NVlabs
ASPIRE
ASPIRE: Agentic /Skills Discovery for Robotics
281 GitHub stars (281)
+55 in 16 days
| Period | New stars |
|---|---|
| Oct 1 | +1 |
| Oct 2 | +1 |
| Oct 3 | +5 |
| Oct 4 | 0 |
| Oct 5 | +2 |
| Oct 6 | +3 |
| Oct 7 | +4 |
| Oct 8 | +3 |
| Oct 9 | +4 |
| Oct 10 | +7 |
| Oct 11 | +3 |
| Oct 12 | +4 |
| Oct 13 | +8 |
| Oct 14 | +2 |
| Oct 15 | +6 |
| Oct 16 | +2 |
- 19 skills (19 skills found in its SKILL.md files)
- 17 forks (17 forks)
- 1d ago last update (Last updated 1d ago)
Install
Skills CLI
npx skills add NVlabs/ASPIRE
Asks which skills and agents to install. Just one: npx skills add NVlabs/ASPIRE --skill yam-full-demo — add -g for every project.
Claude Code
git clone https://github.com/NVlabs/ASPIREcp -r "ASPIRE/aspire/real/.agents/skills/yam-full-demo" ~/.claude/skills/
Copies the yam-full-demo skill into ~/.claude/skills/ for every project (use .claude/skills/ for one project). Repeat for the other skills listed below.
Plugin
/plugin marketplace add NVlabs/ASPIRE
Run inside Claude Code if the repository ships a plugin marketplace, then pick plugins from /plugin.
Skills can run code on your machine — read the SKILL.md first.
What it does
In the author's own words, from each skill's SKILL.md.
-
yam-full-demo
Use for the current YAM full demo, yam_demo.sh, bottle rack, KitKat handover/trash, bowl/can commands, and canonical saved-script commands.
aspire/real/.agents/skills/yam-full-demo/SKILL.md on GitHub (opens in a new tab) -
yam-geometry
Use for YAM DOF, coordinate directions, tabletop/rack/trash geometry, camera frames, gripper geometry, and contact/placement offsets.
aspire/real/.agents/skills/yam-geometry/SKILL.md on GitHub (opens in a new tab) -
yam-grasp-pickup
Use for YAM object detection, grasp candidate generation, side/top-down/rim pickup choices, staged gripper close, contact evidence, lift-only checks, and pickup failures.
aspire/real/.agents/skills/yam-grasp-pickup/SKILL.md on GitHub (opens in a new tab) -
yam-motion-planner
Use for YAM cuRobo planning/IK failures, no-motion vs real-robot planning mismatch, waypoint robustness sweeps, perturbation testing, and recovery after motion planner failures.
aspire/real/.agents/skills/yam-motion-planner/SKILL.md on GitHub (opens in a new tab) -
yam-retreat-recovery
Use for YAM post-release retreat, gripper-open confirmation, home/open recovery, retreat after rack/bin placement, and safe recovery while holding an object after a failed run.
aspire/real/.agents/skills/yam-retreat-recovery/SKILL.md on GitHub (opens in a new tab) -
yam-robot-debugging
Use for YAM real-robot debugging: saved-script edits, physical runs, artifact inspection, autonomous patch-and-retry loops, and compact runtime context updates.
aspire/real/.agents/skills/yam-robot-debugging/SKILL.md on GitHub (opens in a new tab) -
yam-runtime-artifacts
Use when inspecting YAM logs, videos, SAM3 overlays, observations, planner previews, function-call JSON, result files, or failure evidence.
aspire/real/.agents/skills/yam-runtime-artifacts/SKILL.md on GitHub (opens in a new tab) -
yam-server-setup
Use for YAM arm, camera, perception, AnyGrasp, BundleSDF, SAM3, cuRobo, PyRoki, and provider server checks/restarts.
aspire/real/.agents/skills/yam-server-setup/SKILL.md on GitHub (opens in a new tab) -
yam-simulation-transfer
Use when borrowing LIBERO/robosuite strategy ideas for YAM grasping, handover, pouring, placement, rack, or bin tasks.
aspire/real/.agents/skills/yam-simulation-transfer/SKILL.md on GitHub (opens in a new tab) -
yam-transport
Use for YAM held-object transport after pickup, rack/bin approach waypoint ordering, object-relative placement, avoiding shelf/rack collisions, and choosing staged axis moves.
aspire/real/.agents/skills/yam-transport/SKILL.md on GitHub (opens in a new tab) -
behavior/fix-loop
Complete BEHAVIOR-1K ASPIRE protocol: learn skills on seeds 26-35, freeze them, then run isolated fresh-agent adaptation on seeds 1-25.
aspire/sim/.claude/behavior/fix-loop/SKILL.md on GitHub (opens in a new tab) -
libero-evosearch
Run the LIBERO-Pro Fix Loop + Evolutionary Search experiment: iterative candidate search for low-performing tasks, validation selection, and final held-out evaluation.
aspire/sim/.claude/libero/evosearch/SKILL.md on GitHub (opens in a new tab) -
libero-fix-loop
Run the baseline-free LIBERO-Pro Fix Loop: inspect one initial observed scene, generate task-level code, debug failures on seeds 51–65 using traces and keyframes, validate on seeds 1–50, and promote reusable patterns.
aspire/sim/.claude/libero/fix-loop/SKILL.md on GitHub (opens in a new tab) -
libero-inference-time-scaling
Run the LIBERO inference-time scaling experiment: debug Long-Pro tasks at frozen library snapshots, evaluate token-budget checkpoints, and plot Pareto curves.
aspire/sim/.claude/libero/inference-time-scaling/SKILL.md on GitHub (opens in a new tab) -
libero-library-size-scaling
Run the LIBERO library-size scaling experiment: evaluate frozen snapshot skill libraries on LIBERO-Long-Pro and generate success-rate scaling tables/plots.
aspire/sim/.claude/libero/library-size-scaling/SKILL.md on GitHub (opens in a new tab) -
libero-zeroshot-transfer
Run the LIBERO-90 build and zero-shot transfer experiment: build a skill library on LIBERO-90, tag snapshots, then evaluate transfer to LIBERO-Long-Pro.
aspire/sim/.claude/libero/zeroshot-transfer/SKILL.md on GitHub (opens in a new tab) -
robosuite-evosearch
Run Robosuite Fix Loop + Evolutionary Search with K=8 candidate search on seeds 101-125 and selected-code evaluation on seeds 1-100.
aspire/sim/.claude/robosuite/evosearch/SKILL.md on GitHub (opens in a new tab) -
robosuite/fix-loop/SKILL
Master reference for ASPIRE Robosuite experiments. Covers system overview, 7 tasks, setup, running experiments, debugging, full API reference, and all pipeline modes (Fix Loop, Baseline).
aspire/sim/.claude/robosuite/fix-loop/SKILL.md on GitHub (opens in a new tab) -
robosuite/training-law/SKILL
Master reference for ASPIRE Robosuite experiments. Covers system overview, 7 tasks, setup, running experiments, debugging, full API reference, and all pipeline modes (Fix Loop, Baseline).
aspire/sim/.claude/robosuite/training-law/SKILL.md on GitHub (opens in a new tab)
No skill matches that filter.
Review before you install
A skill is a set of instructions and scripts your agent will follow, and it can run code on your machine. Read the SKILL.md and any scripts first — this listing is not a security review.