Choosing QEC Decoders

What it does

This skill helps researchers and engineers select the right quantum error-correction (QEC) decoder backend for a given workload. It frames core routing questions (artifact shape, code family, noise model, desired benchmark behavior) and provides routing defaults and checklists to map problems to decoder families such as matching-based, search-based, or specialized color-code decoders. The guidance prevents misapplication of decoders and encourages documenting constraints and rationale.

When to use it

Use this skill when the decision point is which decoder backend to choose — before implementing adapters or running decoder experiments. Scenarios: picking a baseline for benchmarking, routing Stim detector-model workflows, choosing a color-code-specialized decoder, or deciding when matching-based decoders are inappropriate.

What's included

• Scripts: none detected (has_scripts=false)
• References: the repo includes a references/ directory with API and examples links and cross-references to related SKILLs (pymatching, fusion-blossom, tesseract-decoder).
• Instructions: a checklist for classifying artifacts, consulting a capability matrix, recording constraints, and documenting why a backend was chosen. Lists common failure modes and fixes to avoid incorrect decoder choices.

Compatible agents

Suited for research-assistant agents and tooling that can read repo references and cross-link SKILLs (research-oriented agents, code review assistants, and experimental orchestration helpers).