toolkit-backends
Choose, inspect, and troubleshoot OpenFF Toolkit optional backend wrappers and ToolkitRegistry behavior.
- License 许可
- MIT
- 最近更新
- 2026-09-03
- Stars
- ★ 351
- Forks
- 27
01 前言 Front Matter
- name
- toolkit-backends
- license
- MIT
- metadata
- { "disco-role": "operating" }
- description
- Choose, inspect, and troubleshoot OpenFF Toolkit optional backend wrappers and ToolkitRegistry behavior.
- disable-model-invocation
- true
02 资源文件 3
- 参考文档references/backend-reference.md8.3 KB
- 参考文档references/troubleshooting.md7.5 KB
- 脚本scripts/check_toolkit_backends.py8.6 KB
03 SKILL.md 正文
Toolkit Backends
Use this sub-skill when a task involves OpenFF Toolkit optional cheminformatics backends, wrapper availability, ToolkitRegistry precedence, GLOBAL_TOOLKIT_REGISTRY, toolkit_registry_manager, supported file formats, supported charge methods, or dependency/license failures for RDKit, OpenEye, AmberTools, OpenFF NAGL, or the built-in wrapper.
Start Here
- Run
python scripts/check_toolkit_backends.py --jsonwhen you need environment-specific backend facts. - Read
references/backend-reference.mdto choose wrappers, registries, optional installs, and supported file/charge capabilities. - Read
references/troubleshooting.mdwhen a wrapper is unavailable, a method cannot be resolved, a file format or charge method is unsupported, or OpenEye/RDKit behavior differs.
Routing Boundaries
- Handle backend readiness first: availability, registry construction, wrapper ordering, explicit
toolkit_registry=use, and optional dependency guidance. - Route molecule construction, file conversion, topology loading, force-field parameterization, and chemistry workflows back to the relevant OpenFF Toolkit domain sub-skill after backend selection is settled.
- Do not rely on global registry mutation unless the user specifically needs process-wide behavior; prefer explicit wrapper/registry arguments or
toolkit_registry_managerfor temporary changes.
04 同仓库更多 Skills
passnet-triton-opt
Make a single PassNet Triton kernel fast AND numerically faithful: performance model (when a replacement can win at all), block/grid/warp tuning, autotune policy, launch overhead, and per-op numeric recipes to pass the dtype baseline tolerances. Use when a pass matches and is correct but speedup below expected, or when correctness fails by small numeric margins.
passnet-solve
END-TO-END playbook for solving one PassNet sample: analyze the computation graphs, decide the optimization strategy (what to fuse, what to replace, what to leave alone), drive the iteration loop, and maximize the sample score. This is the ENTRY skill — invoke it first for any PassNet optimization task; it tells you when to use passnet-pattern-fusion, passnet-triton-opt and passnet-feedback.
passnet-skill
PassNet GPU kernel optimization via compiler passes. Design and implement Triton-based optimization passes, create pass files under ./pass_dir/, self-evaluate with pass_evaluator, and iterate to maximize GPU speedup.
passnet-pattern-fusion
Author PassNet pass files: write patterns that actually MATCH the FX graph, pick fusion regions (which ops to absorb into one Triton kernel), and structure multi-pass submissions with the mandatory shared-dispatch architecture. Use when creating or fixing pattern()/replacement_args()/replacement_func() files, when a pass "failed to match", or when deciding how to fuse multiple kernels/ops.
passnet-orchestrate
Round/batch planning and decision support for PassNet graph optimization. Use for multi-sample triage, worker allocation, eval-budget policy, and post-eval keep/revert decisions. For solving one concrete sample end to end, use passnet-solve as the entry skill; consult this skill only when you need broader planning or a specific decision gate.
passnet-feedback
Fast, deterministic feedback for PassNet work: pre-flight pattern/match verification WITHOUT burning a GPU evaluation, per-node bottleneck analysis of a sample's graphs, and eval-log parsing into per-variant status + estimated score + failure classification. Use BEFORE every GPU evaluation (check_pattern), at the START of a sample (analyze_graph), and AFTER every evaluation (parse_eval_log).