Individual skill
setup-af3_torch-remote-tool
Set up, launch, validate, and troubleshoot the AF3-torch ToolUniverse remote tool and optionally relay it through ToolUniverse Connect. Use when deploying or auditing this implementation.
skills/setup-af3_torch-remote-tool/SKILL.md
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.
Standard SKILL.md format has not been attested at this commit. No installation command is available; inspect the source.
Repository context
- Repository stars (not skill ratings)
- 1,779
- Repository last updated
- Nov 8, 2026
- License
- Apache-2.0
Observed agent integration: No specific integration evidence. Standard SKILL.md installation does not imply verified compatibility.
- agents Repository topic
- reasoning-agent Repository topic
- reasoning-language-models Repository topic
- tool-use Repository topic
- ai-agents Repository topic
- ai-scientists Repository topic
- automated-science Repository topic
- co-pilot Repository topic
- co-scientist Repository topic
- ai-for-science Repository topic
- engineering Derived category
Files and excerpt
File metadata and permitted, redacted excerpts come from the recorded scan. Source links are pinned to that commit.
Browse files
SKILL.mdcore · 14,007 bytesSKILL.mdcore · 14,007 bytesSKILL.mdcore · 14,007 bytesSKILL.mdcore · 14,007 bytesSKILL.mdcore · 14,007 bytesSKILL.mdcore · 14,007 bytesSKILL.mdcore · 14,007 bytesSKILL.mdcore · 14,007 bytesSKILL.mdcore · 14,007 bytesSKILL.mdcore · 14,007 bytesSKILL.mdcore · 14,007 bytesSKILL.mdcore · 14,007 bytes
Read excerpt
# Set up AF3-torch as a remote tool
> Validation status (2026-10-01): AlphaFold3 (the xfold PyTorch re-implementation, Apache-2.0, on the public OpenFold3-preview2 checkpoint -- not DeepMind's licence-gated parameters) ran on the NVIDIA GB10 (aarch64, cc 12.1) under the optimization kit's `fast` mode. This deployment was expected not to be possible on aarch64 and is: the one C++ extension its second environment needs built clean over 236 targets with no portability patch, and the converted weights file is bit-exact against a sha256 the kit measured on x86-64. Verified on FOUR CONSECUTIVE IDENTICAL REQUESTS plus a mixed-shape sequence and an ascending ladder: forward 3.51 s cold then 1.46 s warm, peak reserved 1.75 GiB at 100 residues rising to 9.21 at 800 -- the steepest ladder of any provider here (5.3x), because the AF3 pair track dominates with no large constant term to flatten it. **This provider needs TWO pinned interpreters**, one for the model and one CPU-only for featurisation and the mmCIF writers; `[env]` declares one, so the second is supplied through environment variables and the weights step refuses by name without it. Public publication, cross-user isolation, concurrency, recovery and structural accuracy 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