Menu

Individual skill

setup-ef2inv-remote-tool

0upvotes

Set up, launch, validate, and troubleshoot the ESMFold2-binder-design ToolUniverse remote tool and optionally relay it through ToolUniverse Connect. Use when deploying or auditing this implementation.

skills/setup-ef2inv-remote-tool/SKILL.md

Read pinned source

Source commit 5598a9fc97e32964fe9b703818e7d0747142f179

What was checked

Matches the observed repository HEAD
Yes
Scanner rules are current
Yes
Repository scan complete
No
Skill scan complete
No
Observed behavior
Network or broad shell
Last checked
Nov 9, 2026

These are observations at the recorded commit, not a guarantee of safety. Review the source before running it.

Install this skill

Choose an agent in your Stack or copy a command below. Commands fetch the pinned commit and copy only this skill directory. Existing destinations are refused. Copying a command does not install anything.

An install command is unavailable: current rules, complete repository and skill scans, an eligible behavior level and a permitted license are required.

Repository context

Repository stars (not skill ratings)
1,780
Repository last updated
Nov 9, 2026
License
Apache-2.0

Observed agent integration: No specific integration evidence. Standard SKILL.md installation does not imply verified compatibility.

Files and excerpt

File metadata and permitted, redacted excerpts come from the recorded scan. Source links are pinned to that commit.

Browse files
Read excerpt

# Set up ESMFold2 binder design as a remote tool

> Validation status (2026-10-01): the ESM cookbook's gradient-guided binder design ran on the NVIDIA GB10 (aarch64, cc 12.1) under the optimization kit's `fast` mode. **This is the only provider here that runs BACKWARD passes**, so the acceptance question is not only whether a graph replay survives a repeat but whether autograd state accumulates across requests. It does not: four consecutive identical requests in one held worker moved the high-water mark by 0.04 GiB, at 15.7 s then 5.1 s each. Peak reserved 28.03 GiB for the 135-token warm pass. **No VRAM ladder is declared, deliberately**: under `fast` the kit's checkpointing planner budgets 0.90x the card's total memory and grows to fill whatever card it is on above roughly 500 tokens, so a ladder would measure the host rather than the model. `shareable = false` says the same thing in the field a scheduler reads, and the bounded alternative is the kit's `big` mode rather than a knob. Public publication, cross-user isolation, concurrency, recovery and design quality are untested; keep this deployment private.

## Prerequisites

- Run from the ToolUniverse repository root on Linux with Python 3.12.3.
- A CUDA GPU is required. This provider holds its model resident between requests; CPU is not a target.
- Keep provider data, weights, caches, and credentials outside Git.
- Bind to loopback. A non-loopback bind requires TOOLUNIVERSE_API_TOKEN; never put it in arguments or results.

Run the standard-library contract check before downloading large dependencies:

~~~bash
Back to top