AGENT SKILL

project-workflows

Operate spaCy project.yml workflows safely with local validation, dry runs, and explicit network boundaries.

License 许可
MIT
最近更新
2026-09-03
Stars
★ 351
Forks
27

01 前言 Front Matter

name
project-workflows
license
MIT
metadata
{ "disco-role": "operating" }
description
Operate spaCy project.yml workflows safely with local validation, dry runs, and explicit network boundaries.
disable-model-invocation
true

02 资源文件 4

03 SKILL.md 正文

project-workflows

Use this sub-skill for project-level spaCy orchestration: cloning templates, fetching assets, running commands and workflows, generating docs, generating DVC config, and pushing or pulling outputs from remotes.

Read or run these first

Default operating order

  1. Validate the project file locally.
  2. Inspect or generate documentation.
  3. Use project run --dry before the first real run.
  4. Fetch assets only when you know whether --extra is needed.
  5. Run the narrowest command that covers the change.
  6. Use push, pull, and dvc only after remotes or DVC are confirmed.

Route elsewhere when the work shifts

  • If the project command centers on training config, spacy train, spacy convert, spacy evaluate, or spacy package, use training-and-cli.
  • If the issue is package import, installation, or model download, use install-and-inspect.
  • If a project script defines custom pipeline components or factories, use pipeline-components.

Verification focus

  • project-document-dry-run: local temp project, project document, and project run --dry.
  • project-assets-remote: help-only and skip-network checks for project assets, push, and pull.

Boundaries

  • Do not treat external template cloning as a validation step.
  • Do not assume remote storage, credentials, or network access are available.
  • Do not move into training internals or model-serving integrations here.

04 同仓库更多 Skills

VectorSpaceLab
VectorSpaceLab/AREX-Skill

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.

★ 351
VectorSpaceLab
VectorSpaceLab/AREX-Skill

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.

★ 351
VectorSpaceLab
VectorSpaceLab/AREX-Skill

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.

★ 351
VectorSpaceLab
VectorSpaceLab/AREX-Skill

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.

★ 351
VectorSpaceLab
VectorSpaceLab/AREX-Skill

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.

★ 351
VectorSpaceLab
VectorSpaceLab/AREX-Skill

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).

★ 351