Temple Compute

A pure-observer plugin (task + workflow middleware, no engine changes) that persists what the engine already computes but does not keep. It answers questions that come after a run: which runs used an input that turned out bad, what consumed their outputs, what must be re-run or blocked. Failures inside it only log, never fail the run.

Run reproducible scientific workflows anywhere (laptop, remote, or HPC cluster) from a single YAML file. The open-source runtime behind Temple Compute.

Horus SLURM executor plugin

Documentation for the Horus Workflow Manager

Production-ready scientific workflows for the Horus Runtime: molecular dynamics, virtual screening, docking, LLM training and more. HPC pipelines that actually run at scale.

Run Jupyter Notebooks as Horus tasks

Use docker as an Horus executor

Horus python (uv/conda) environment auto-provisioning for tasks.

Template repository for horus-runtime plugins.