Token导航 LogoToken导航TokenDH.com
研究检索external-servicegithub未标认证来源可访问许可证需确认审计异常

lifecyclemodel-resulting-process-builder生命周期模型结果流程构建器

Agent Skill

lifecyclemodel-resulting-process-builder 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

总安装

186

周安装

8

GitHub Stars

4

下载量

65
CodexClaudeCursorGemini CLI

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

2

许可证

unknown

最后核验

2026-05-01

来源状态

来源可访问

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:lifecyclemodel-resulting-process-builder(生命周期模型结果流程构建器)
来源仓库:https://github.com/tiangong-lca/skills
仓库路径:skills/lifecyclemodel-resulting-process-builder
安装命令:
npx skills add https://github.com/tiangong-lca/skills --skill lifecyclemodel-resulting-process-builder
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 npx skills 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

skills.shnpx skills
npx skills add https://github.com/tiangong-lca/skills --skill lifecyclemodel-resulting-process-builder

简介

lifecyclemodel-resulting-process-builder 用于查找、检索和筛选相关信息。

  • 适合在 Codex、Claude、Cursor、Gemini CLI 中根据关键词快速定位候选结果。
  • 通过 npx skills add 命令从指定仓库安装,需确认权限和维护状态。
  • 使用前建议核验是否会触发联网、命令执行或文件读写操作。
  • 适用宿主包括 Codex、Claude、Cursor、Gemini CLI,接入前应确认版本、权限和运行环境要求。

SKILL.md

Lifecycle Model Resulting Process Builder

Use this skill when the source of truth is already a lifecycle model json_ordered file and the next step is to deterministically compute the aggregated resulting process plus relation handoff artifacts, not to synthesize a process from external flow evidence.

Run Workflow

  1. By default the wrapper runs the published CLI through npm exec --yes --package=@tiangong-lca/cli@latest -- tiangong. Use TIANGONG_LCA_CLI_DIR or --cli-dir only for local dev/CI overrides.
  2. Use node scripts/run-lifecyclemodel-resulting-process-builder.mjs build... to delegate to tiangong lifecyclemodel build-resulting-process.
  3. Use node scripts/run-lifecyclemodel-resulting-process-builder.mjs publish... to delegate to tiangong lifecyclemodel publish-resulting-process.
  4. Confirm the local artifacts in the run directory before any later tiangong publish run step.

The active runtime path is skill -> Node wrapper -> tiangong CLI. Python and MCP are no longer part of the normal execution path for this skill.

What The Implementation Does

  • validates assets/request.schema.json
  • loads a lifecycle model from source_model.json_ordered or source_model.json_ordered_path
  • resolves referenced process datasets from local process exports
  • extracts process instances and graph edges from model topology
  • aggregates exchanges across included processes and cancels internal linked flows
  • derives:

- process-projection-bundle.json - projection-report.json - publish-bundle.json via publish - publish-intent.json via publish

  • emits relation payloads containing:

- generated_from_lifecyclemodel_id - generated_from_lifecyclemodel_version - projection_role - projection_signature

  • supports primary-only
  • accepts all-subproducts requests conservatively and reports when only a primary aggregated process can be emitted
  • keeps all work local; no remote write path is executed here

Inputs

Always provide:

  • source_model
  • projection
  • publish

Provide process_sources when local process resolution is not discoverable from the model path.

The source model may be provided as:

  • source_model.id
  • source_model.json_ordered_path
  • source_model.json_ordered

Referenced process datasets may be provided via:

  • process_sources.process_catalog_path
  • process_sources.run_dirs[]
  • process_sources.process_json_dirs[]
  • process_sources.process_json_files[]
  • auto-detected sibling directories such as processes/ or *-processes/ when using --model-file

Canonical request files should use process_sources.allow_remote_lookup, but the normal skill flow is still local-first and should keep it false unless deterministic remote process lookup is explicitly needed. When process_sources.allow_remote_lookup=true, supply TIANGONG_LCA_API_BASE_URL and TIANGONG_LCA_API_KEY or pass --base-url and --api-key through to the CLI.

Outputs

  • request.normalized.json
  • source-model.normalized.json
  • source-model.summary.json
  • projection-report.json
  • process-projection-bundle.json
  • publish-bundle.json from publish
  • publish-intent.json from publish

Commands

node scripts/run-lifecyclemodel-resulting-process-builder.mjs build \
  --request assets/example-request.json \
  --out-dir /abs/path/run-001

node scripts/run-lifecyclemodel-resulting-process-builder.mjs build \
  --model-file assets/example-model.json \
  --projection-role primary \
  --out-dir /abs/path/run-001

node scripts/run-lifecyclemodel-resulting-process-builder.mjs publish \
  --run-dir /abs/path/run-001 \
  --publish-processes \
  --publish-relations

# Force a local CLI working tree during dev/CI
TIANGONG_LCA_CLI_DIR=/path/to/tiangong-lca-cli \
  node scripts/run-lifecyclemodel-resulting-process-builder.mjs build --json

Use --model-file only when local process sources can be inferred from the model location. Use --request to pin process_sources.* explicitly. The wrapper keeps --request and --model-file as compatibility flags, but the underlying CLI contract is --input <request.json> for build and --run-dir <dir> for publish.

Separation Rule

  • use process-automated-builder for flow-to-process synthesis
  • use this skill for lifecyclemodel-to-resulting-process aggregation/build

They are different pipelines and should stay separate.

Load References On Demand

  • references/projection-workflow.md: intake to publish-handoff stages
  • references/projection-contract.md: bundle shape and relation semantics
  • references/builder-invocation-contract.md: caller/callee contract
  • references/integration-notes.md: current cross-project architecture notes

适合场景

01

用户想查找某类 Agent Skill 时

02

需要根据任务场景推荐可安装能力包时

03

需要对比不同来源的安装命令和来源信息时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

Codex

39.81%
按下载量换算26

Claude

31.75%
按下载量换算21

Cursor

16.99%
按下载量换算11

Gemini CLI

9.25%
按下载量换算6

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

未通过

权限和风险

external-service

该 Skill 可能调用第三方服务、云服务或外部模型 API,使用前需要确认账号、额度、数据发送范围和服务条款。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。来源安全扫描存在 warning/failed 结果,不能写成本站确认安全。当前只有一个来源,正式发布前建议补源仓库或其他目录站核验。

来源信息

继续浏览同类 Skills