fine-tuning-openvla-oft

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.

Quality

100%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Quality

Content

100%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The content is lean and highly actionable, with clear sequenced workflows, validation checkpoints, and well-signaled references to real bundle files. It balances breadth with token efficiency and progressive disclosure.

Dimension	Reasoning	Score
Conciseness	The body is dense with executable commands, tables, and checklists with minimal padding; concept framing (e.g. OFT vs tokenized actions) is brief and conveys skill-specific knowledge rather than general background Claude already has.	3 / 3
Actionability	Provides fully executable, copy-paste-ready commands (torchrun finetune, deploy.py, run_libero_eval.py) plus a concrete log-parsing function, not pseudocode or vague direction.	3 / 3
Workflow Clarity	Each workflow has a progress checklist and numbered steps, plus a 'Critical invariants' table and a config-parity validation snippet with issue→fix feedback loops in 'Common issues'.	3 / 3
Progressive Disclosure	SKILL.md is a well-organized overview that defers detail to one-level-deep references (aloha-workflow.md, libero-workflow.md, paper-and-checkpoints.md, config-troubleshooting.md), all of which exist in references/.	3 / 3
	Total	12 / 12 Passed

Description

100%

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description is specific, third-person, and well-scoped to a distinct OpenVLA-OFT niche. It cleanly answers both what the skill does and when to invoke it with natural trigger terms.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions ('Fine-tunes and evaluates OpenVLA-OFT and OpenVLA-OFT+ policies', 'continuous action heads, LoRA adaptation, and FiLM conditioning on LIBERO simulation and ALOHA real-world setups') rather than vague language.	3 / 3
Completeness	Explicitly answers both what it does and when to use it via a clear 'Use when reproducing... training... deploying... or debugging...' trigger clause.	3 / 3
Trigger Term Quality	Good coverage of natural terms users would say in this domain — 'OpenVLA-OFT', 'LIBERO', 'ALOHA', 'LoRA', 'server-client inference', 'normalization' — matching the vocabulary of the target audience.	3 / 3
Distinctiveness Conflict Risk	Tied to a clear niche (OpenVLA-OFT/OFT+ specifically) with distinct triggers, making it unlikely to fire for unrelated skills.	3 / 3
	Total	12 / 12 Passed

Validation

93%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: Orchestra-Research/AI-Research-SKILLs
Path: 18-multimodal/openvla-oft/SKILL.md
Commit: 773a529

Reviewed: about 11 hours ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.