Traction scoreGitHub stars can be faked, so popularity alone can be misleading. Traction Score looks for broader signs of recent attention, adoption, and active maintenance.
Compatible AgentsThe repository documents support for these agents. The skills may also work with other agents that can load SKILL.md files, but they may need some setup or small changes.
Follow the documented setup, then try a first task.
I’m using
ASPIRE simulation Quick Start with a coding agent
Installs: Clone the main ASPIRE repository, including its simulation skills, and prepare the complete ten-task LIBERO-Pro Goal-Swap Quick Start. The documented prompt requires confirmation before launching. · Codex, Claude Code
Before you start
Host setup completed using docs/host-setup.md, including CUDA, uv, and a Hugging Face token for gated SAM3 weights.
An isolated execution host without sensitive credentials or mounts, restricted network access, and no simulation-agent access to physical hardware.
Launch a coding agent such as Codex or Claude Code.
Enter this complete prompt in the agent session, not a terminal. Review the reported requirements before confirming the benchmark launch.
Clone the main branch of https://github.com/NVlabs/ASPIRE/,
read the iterative-debugging skill (.claude/skills/iterative-debugging/SKILL.md) and
docs/libero-pro.md, and run the complete ASPIRE LIBERO-Pro Goal-Swap Quick Start for
all ten tasks in the libero_goal_swap suite.
Before executing, report the required GPUs, credentials, gated weights,
services, expected runtime, seed partitions, and output paths. Wait for
my confirmation before launching. Do not access real-robot code or instructions
(AGENTS.md, docs/system2*.md, docs/yam-real.md, src/cap_harness/yam_real/).
An ASPIRE repository checkout containing the referenced benchmark guides and coordinator runbook.
Host and benchmark setup details, including GPU availability, pinned LIBERO environment, SAM3 authentication, and perception-service status.
Use the iterative-debugging skill to prepare a campaign for LIBERO-Pro, suite libero_object_swap, task_id 0. Before launching any episodes, report the required setup, development and evaluation seed partitions, services, and intended output paths. Identify missing prerequisites and stop after this preparation report.
One low-performing task identified by benchmark, suite, and task_id.
An ASPIRE checkout and the target benchmark's setup details.
The task's existing fix_code.py and baseline held-out rate, when available.
Use the evosearch skill to review readiness for one low-performing benchmark task I provide. Check benchmark setup, SAM3 authentication, perception services, and any existing fix_code.py and baseline held-out rate. Explain how candidate_A and the development and held-out partitions apply, then list missing inputs. Stop before dispatching workers or running evaluations.
An existing ASPIRE CAP session with native CAP MCP tools and a calibrated RGB-D view.
The repository's AGENTS.md operational rules.
Make sure cap-tool-usage is installed in your agent before trying this task.
Use the cap-tool-usage skill to localize the object I identify using the existing ASPIRE CAP session. Discover the current MCP schemas and available camera names, use a calibrated RGB-D view, and report the object's identity, frame, geometry, and localization diagnostics. If the subject is ambiguous, inspect an available view to disambiguate it. Keep this task read-only.
Claude
Codex