An experiment workflow for LIBERO-Pro, a robot-learning benchmark, that searches for fixes to tasks that perform poorly and tests the selected fixes.

Use it to coordinate per-task search iterations, dispatch work to available GPUs, reset tasks before reruns, and evaluate results on held-out tasks.

It organizes repeated candidate generation, validation, reruns, and final testing so experiments do not depend on an improvised process.

How NVlabs/ASPIRE configures Claude Code

Reuse

Borrowing it

Nothing to install: this file belongs to NVlabs/ASPIRE. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file

curl -O https://raw.githubusercontent.com/NVlabs/ASPIRE/main/.claude/skills/evosearch/SKILL.md

Clone the repo

git clone --depth 1 https://github.com/NVlabs/ASPIRE

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

README.md

[![agentmods](https://agentmods.dev/badge/skills/nvlabs/aspire/evosearch/github.svg)](https://agentmods.dev/skills/nvlabs/aspire/evosearch)

Your own site

<a href="https://agentmods.dev/skills/nvlabs/aspire/evosearch"><img src="https://agentmods.dev/badge/skills/nvlabs/aspire/evosearch/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/skills/nvlabs/aspire/evosearch"><img src="https://agentmods.dev/badge/skills/nvlabs/aspire/evosearch.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 57 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 781 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits

NVIDIA SkillSpector pass 7 Sept 2026

How audits are shown

Origin original No closer match found in the catalogue.

Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

Model	Per session	Once invoked
Fable 5.1	$0.00057	$0.00781
Opus 5.5	$0.00023	$0.00312
Sonnet 5.5	$0.00011	$0.00156
Haiku 4.5	$0.00006	$0.00078

Measured 21d ago against content hash 7f7f4b86a35c, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

evosearch scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 21d ago.

The scan reads SKILL.md. This mod also ships 2 executable files (scripts/analyze_evosearch_traces.py, scripts/evosearch_eval.py), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.claude/skills/evosearch/SKILL.md · 54 lines

How it starts

The opening of the file, as written. The whole thing — 54 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Fix Loop + Evolutionary Search

Use this skill for intensive debugging of difficult tasks — Evolutionary Search's multi-candidate search finds strategies the single-pass fix loop misses. It builds on the iterative-debugging skill: each task's fix loop fix_code.py (when present) is the baseline to beat, seeded verbatim as candidate_A. LIBERO-Pro runs use --init-mode seeded so development seeds 51–65 and held-out seeds 1–50 are distinct scenes.

Setup

Complete benchmark setup for the target embodiment (docs/libero-pro.md, or docs/robosuite.md), including the pinned environment (for LIBERO: scripts/bootstrap_libero.sh → .venv-libero) and SAM3 authentication.
Export the benchmark environment variables (for LIBERO: LIBERO_CONFIG_PATH="$PWD/.libero", MUJOCO_GL=egl, PYOPENGL_PLATFORM=egl, EGL_PLATFORM=device, TORCH_FORCE_NO_WEIGHTS_ONLY_LOAD=1).
Start perception services (scripts/supervise_services.sh) and a persistent coordinator session.
Confirm the Fix Loop experiment has produced fix_code.py (and a baseline held-out rate) for the target tasks when available.
Read docs/api-reference.md, docs/run-artifacts.md, and main-agent-prompt.md.

Run Order

Take the task list from the user (tasks below the success threshold; any set of (benchmark, suite, task_id) cases) and follow main-agent-prompt.md as the coordinator.
Fill subagent-prompt.md once per task and dispatch subagents on the available worker GPUs.
Subagents read the experiment-specific companion skills under skills/ when generating candidates, iterate with scripts/evosearch_eval.py, and run the Stage 2 held-out evaluation themselves with the shared .claude/skills/iterative-debugging/scripts/run_validation.py before returning.
Use clean-task-slate.md before rerunning a task or suite.

Read the full file on GitHub · 54 lines

Files

What ships with it

7 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

21d ago Changed · +24 lines · +20 tokens per session 7f7f4b86a35c
2mo ago First seen · 30 lines · 37 tokens per session scan A 5357f33e771e

Subscribe to this mod's changes

evosearch is a skill published in the GitHub repository NVlabs/ASPIRE (959 stars, last pushed 22d ago), licensed Apache-2.0. It adds 57 tokens to every session and 781 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Scientific Testing

evosearch next to 6 skills from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
evosearchNVlabs/ASPIRE	Skill	959	22d ago	Apache-2.0	A	57	Claude Code
instrument-data-to-allotropeanthropics/knowledge-work-plugins	Skill	35k	2d ago	Apache-2.0	A	123	Claude Code
exploratory-data-analysisK-Dense-AI/scientific-agent-skills	Skill	56k	4d ago	MIT	A	83	Claude Code
matlabK-Dense-AI/scientific-agent-skills	Skill	56k	4d ago	MIT	A	44	Claude Code
phylogeneticsK-Dense-AI/scientific-agent-skills	Skill	56k	4d ago	MIT	A	71	Claude Code, Codex
research-engineerdavila7/claude-code-templates	Skill	33k	yesterday	MIT	A	43	Claude Code, Codex
mapping-to-snomedmaziyarpanahi/openmed	Skill	5.8k	yesterday	Apache-2.0	A	205	Claude Code

ASPIRE: Skill for Claude Code