# Lifecyclemodel Automated Builder > Assemble native TianGong TIDAS lifecyclemodel `json_ordered` artifacts from existing local process-build runs through the unified TianGong CLI. Use when you already have `process-automated-builder` outputs and need the supported lifecyclemodel auto-build workflow. - Skill: `tiangong-lca/lifecyclemodel-automated-builder` (Agent Skill, multi-file: 8 files) - Install (CLI): `npx skillmds@latest add tiangong-lca/lifecyclemodel-automated-builder` - Raw SKILL.md: https://api.skillmd.com/api/skills/tiangong-lca/lifecyclemodel-automated-builder/raw - Safety review: pending - Works with: Claude Code, Claude.ai, OpenAI Codex - Category: Productivity - Author: tiangong-lca (https://skillmd.com/u/tiangong-lca) - Updated: 2026-09-21 - Page: https://skillmd.com/skills/tiangong-lca/lifecyclemodel-automated-builder --- # Lifecycle Model Automated Builder Use this skill when the source of truth is a set of existing local `process-automated-builder` run directories and the next step is to assemble a native lifecyclemodel artifact locally. ## Read First 1. `references/workflow.md` 2. `references/model-contract.md` 3. `references/source-analysis.md` ## Guardrails - The canonical runtime path is `skill -> Node wrapper -> tiangong-lca CLI`. - Persistent build outputs must use an explicit `--out-dir`; this skill does not choose a default output root. - For repeatable runs, use an explicit output directory such as `/abs/path/artifacts//...`. - The current canonical slices are: - `tiangong-lca lifecyclemodel auto-build` - `tiangong-lca lifecyclemodel validate-build` - `tiangong-lca lifecyclemodel publish-build` - The supported workflow is CLI-based and local to your build inputs: - no Python workflow - no MCP transport - no remote lifecyclemodel CRUD - no reference-model discovery against KB / LLM services - The skill produces native `json_ordered` only. It does not emit `json_tg`, `rule_verification`, or resulting-process artifacts. - `validate-build` and `publish-build` now exist as dedicated CLI follow-up commands; do not reintroduce those stages inside the skill. - Only `local_runs[]` is executable today. Discovery hints may be recorded as deferred notes, but they are not executed inside this skill. ## Workflow 1. Prepare a manifest whose core input is `local_runs[]`. 2. Pick an output directory, typically under a path such as `/abs/path/artifacts//...`. 3. Run `node scripts/run-lifecyclemodel-automated-builder.mjs build --input --out-dir `. 4. During assembly, preserve TianGong native model conventions from `tiangong-lca-next`: - `lifeCycleModelInformation.quantitativeReference.referenceToReferenceProcess` - `technology.processes.processInstance[*].referenceToProcess` - `technology.processes.processInstance[*].connections.outputExchange` - computed `@multiplicationFactor` - a valid `referenceToResultingProcess` reference inside `json_ordered` 5. Review the local outputs: - `run-plan.json` - `resolved-manifest.json` - `selection/selection-brief.md` - `discovery/reference-model-summary.json` - `models/**/tidas_bundle/lifecyclemodels/*.json` - `models/**/summary.json` - `models/**/connections.json` - `models/**/process-catalog.json` - `reports/lifecyclemodel-auto-build-report.json` 6. If the workflow later needs validation or publish handoff, call the dedicated CLI follow-up commands instead of rebuilding those paths inside the skill: - `node scripts/run-lifecyclemodel-automated-builder.mjs validate --run-dir ` - `node scripts/run-lifecyclemodel-automated-builder.mjs publish --run-dir ` 7. If someone asks for remote discovery or AI-assisted model selection, add it as a native `tiangong-lca lifecyclemodel ...` capability first. ## Commands ```bash node lifecyclemodel-automated-builder/scripts/run-lifecyclemodel-automated-builder.mjs build \ --input lifecyclemodel-automated-builder/assets/example-request.json \ --out-dir /abs/path/artifacts//lifecyclemodel-auto-build \ --dry-run node lifecyclemodel-automated-builder/scripts/run-lifecyclemodel-automated-builder.mjs build \ --input /abs/path/request.json \ --out-dir /abs/path/artifacts//lifecyclemodel-auto-build node lifecyclemodel-automated-builder/scripts/run-lifecyclemodel-automated-builder.mjs build \ --input lifecyclemodel-automated-builder/assets/example-local-runs.json \ --out-dir /abs/path/artifacts//lifecyclemodel-auto-build node lifecyclemodel-automated-builder/scripts/run-lifecyclemodel-automated-builder.mjs validate \ --run-dir /abs/path/artifacts//lifecyclemodel-auto-build node lifecyclemodel-automated-builder/scripts/run-lifecyclemodel-automated-builder.mjs publish \ --run-dir /abs/path/artifacts//lifecyclemodel-auto-build ``` ## Troubleshooting - Local CLI override issues: set `TIANGONG_LCA_CLI_DIR` or pass `--cli-dir` only when you intentionally need an unpublished working tree. - Missing `--out-dir`: the wrapper requires an explicit output path such as `/abs/path/artifacts//...`. - Missing `local_runs[]`: the current canonical slice only accepts local process-build runs. - Validation/publish follow-up: use the dedicated CLI subcommands against one existing auto-build run; they work from local build outputs and do not perform remote writes. - Validation failures on required model fields: inspect `references/model-contract.md`. - Topology disagreements: inspect `references/source-analysis.md` for native lifecycle model conventions. - If you need remote discovery or writes, add that capability to the native CLI instead of extending this skill with a separate runtime. ## Bundled Resources - `scripts/run-lifecyclemodel-automated-builder.mjs`: native Node wrapper that delegates to `tiangong-lca lifecyclemodel ...`. - `assets/example-request.json`: minimal current-slice manifest using `local_runs[]`. - `assets/example-local-runs.json`: multi-run local assembly manifest example. - `references/workflow.md`: current CLI-backed workflow and deferred slices. - `references/source-analysis.md`: extracted conventions from `tiangong-lca-next`, `tidas-sdk`, and `tidas-tools`. - `references/model-contract.md`: native `json_ordered` fields required before validation or publish.