界面已本地化;技能介绍与说明文档按所选语言由 AI 自动翻译,代码、命令、路径和专业术语保留原文,以原作者文档为准。
探索技能
效率与协作 · SKILL.md

component-cli

component-cli

【效率与协作】Guides SecretFlow component CLI, component evaluation payloads, model export, serving inferencer, and plugin packaging workflows.

独立 GitHub 快照 · 自动分类 · 以原作者文档为准

技能说明

GitHub

name: component-cli description: "Guides SecretFlow component CLI, component evaluation payloads, model export, serving inferencer, and plugin packaging workflows." disable-model-invocation: true metadata: disco-role: operating license: Apache 2.0

Component CLI and export

Use this sub-skill when the task is driven by the secretflow component CLI or by component evaluation/export payloads rather than by direct model classes.

Owns

  • secretflow component command flows
  • component registry inspection and translation
  • comp_eval, Registry, NodeEvalParam, StorageConfig, SFClusterConfig
  • component-driven model export and serving model packaging
  • plugin entry points and component registration patterns
  • input/output validation for component payloads

Does not own

  • device/object creation or federated dataframe setup — use runtime-data
  • direct preprocessing/statistics/classical ML APIs — use analytics
  • PSI, Kuscia, TEEU, or deployment orchestration — use privacy-orchestration

Trigger phrases

Use this route when a user asks things like:

  • how to list or inspect SecretFlow components
  • how to run a component from a NodeEvalParam payload
  • how to export a trained component pipeline into a serving package
  • how to build or debug a plugin entry point
  • why a component id or DistData payload is rejected

Reading order

  1. Read references/component-cli.md for the command map and payload flow.
  2. Read the root troubleshooting page if the install, protobuf, or entry-point path is broken.
  3. Use scripts/component_registry_smoke.py when you want a tiny no-network check that the component registry is alive.
  4. Use scripts/plugin_entry_template.py as the small reusable plugin entry skeleton when the task is about packaging a plugin.

Workflow

  1. Decide whether the question is about the CLI surface itself or about the component payloads underneath it.
  2. If the task is payload-driven, verify the input/output order first. Most component failures here are ordering or metadata problems, not algorithm bugs.
  3. If the task is an export or serving task, ensure the input pipeline and the model component are ordered exactly as the export logic expects.
  4. Use the registry smoke helper only to prove that the installation is alive; do not treat it as a substitute for a full component payload.

Common decisions

  • Use component ls when you need the live registry surface.
  • Use component inspect --all when you need the public component definitions.
  • Use component translate when you are working on plugin/component metadata.
  • Use component run only when you already know the NodeEvalParam, storage, and cluster inputs are valid.

Bundled files

  • references/component-cli.md — CLI map, payload flow, and export notes.
  • scripts/component_registry_smoke.py — tiny registry import/check helper.
  • scripts/plugin_entry_template.py — safe plugin entry skeleton adapted from the example plugin.
我的收藏