Waxell

skill

fine-tuning-openvla-oft Review findings

orchestra-researchorchestra-research/ai-research-skillsupdated

Browse GitHub Download zipBook a demo
Fine-tunes and evaluates OpenVLA-OFT and OpenVLA-OFT+ policies for robot action generation with continuous action heads, LoRA adaptation, and FiLM conditioning on LIBERO simulation and ALOHA real-world setups. Use when reproducing OpenVLA-OFT paper results, training custom VLA action heads (L1 or diffusion), deploying server-client inference for ALOHA, or debugging normalization, LoRA merge, and cross-GPU issues.
orchestra-research's own description, from the skill's files.
ResearchEngineeringLegal & complianceskill
Stars
14,234
Version
v1
Copies
0
OverviewSecurity7Versions1

Overview

holdReview findings

Waxell's checks found 7 findings to review, highest severity high.

Highest severity
high
Findings
7
What ran
Static analysis

A clean result means nothing alarming in what is published, not a guarantee. How the checks work

We don't republish the skill's instructions. Read them on GitHub at the reviewed commit.

Security

Security scan

Checked 2 days ago

Result

Review findings Waxell's checks found 7 findings to review, highest severity high.

Checked with static analysis: servers it launches, packages it installs at runtime, scripts piped to a shell and bundled binaries. Nothing was run. How the checks work

Findings (7)

high · 4

  • high

    Tells Claude to install unpinned packages (packaging, ninja)

    SKILL.md:118 · skill_risk:S2

  • high

    Tells Claude to install unpinned packages (torch, torchvision, torchaudio)

    references/aloha-workflow.md:18 · skill_risk:S2

  • high

    Tells Claude to install unpinned packages (uvicorn, fastapi, json-numpy)

    references/aloha-workflow.md:20 · skill_risk:S2

  • high

    Tells Claude to install unpinned packages (torch, torchvision, torchaudio)

    references/libero-workflow.md:18 · skill_risk:S2

medium · 3

  • medium

    Tells Claude to install packages (torch==2.2.0, torchvision==0.17.0, torchaudio==2.2.0)

    SKILL.md:105 · skill_risk:S3

  • medium

    Tells Claude to install packages (robosuite==1.4.0)

    SKILL.md:106 · skill_risk:S3

  • medium

    Tells Claude to install packages (flash-attn==2.5.5)

    SKILL.md:119 · skill_risk:S3

Versions

Version history

Each distinct content we have seen, newest first, with the verdict it got at the time.

One version so far: v1 (1.0.0), seen 2026-11-03.

See and control which skills your people use

Inside your organization, Waxell scans each skill and plugin before anyone installs it, approves what passes and publishes it to a private marketplace, and puts every MCP tool call behind one governed URL.

Book a demo

30 minutes. No commitment.

SOC 2 Type II attested · ISO/IEC 27001:2022 certified · HIPAA with a BAA · aligned to the EU AI Act and NIST AI RMF

More research skills and plugins

See all