Skills caixu-build-asset-library

Name: caixu-build-asset-library
Author: openclaw

Build Personal Asset Library. Use when the user wants to convert parsed materials into asset cards and a queryable library at the asset-build stage, including “建资产库”“生成资产卡”“去重整理版本”. Prefer caixu-skill when the user asks for the full end-to-end mainline or is unsure which stage to run. This skill creates a build_asset_library pipeline run, reads parsed files from caixu-data-mcp, extracts canonical asset cards from trustworthy parsed text, merges likely duplicate versions conservatively, persists assets and merged groups, and completes the pipeline run with library_id. Do not use it for querying, lifecycle judgment, package creation, or submission.

install

source · Clone the upstream repo

git clone https://github.com/openclaw/skills

Claude Code · Install into ~/.claude/skills/

T=$(mktemp -d) && git clone --depth=1 https://github.com/openclaw/skills "$T" && mkdir -p ~/.claude/skills && cp -r "$T/skills/angjustinl/asset-library-skill/caixu-build-asset-library" ~/.claude/skills/openclaw-skills-caixu-build-asset-library && rm -rf "$T"

OpenClaw · Install into ~/.openclaw/skills/

T=$(mktemp -d) && git clone --depth=1 https://github.com/openclaw/skills "$T" && mkdir -p ~/.openclaw/skills && cp -r "$T/skills/angjustinl/asset-library-skill/caixu-build-asset-library" ~/.openclaw/skills/openclaw-skills-caixu-build-asset-library && rm -rf "$T"

manifest: skills/angjustinl/asset-library-skill/caixu-build-asset-library/SKILL.md

Build Personal Asset Library

在用户要“建资产库”“生成资产卡”“去重整理版本”时使用这个 skill。

Quick flow

读取已解析文件
先做 document triage，再抽取
```
asset_card
```
保守归并并写入资产库

Required tools

```
caixu-data-mcp.create_pipeline_run
```
```
caixu-data-mcp.get_parsed_files
```
```
caixu-data-mcp.append_pipeline_step
```
```
caixu-data-mcp.upsert_asset_cards
```
```
caixu-data-mcp.upsert_merged_assets
```
```
caixu-data-mcp.complete_pipeline_run
```

Required input

```
library_id
```
```
file_ids[]?
```

Workflow

先创建
```
build_asset_library
```
pipeline run，并在关键阶段追加
```
append_pipeline_step
```
。
调用
```
get_parsed_files
```
，优先限定到当前批次的
```
file_ids[]
```
。
跳过没有可信文本的文件；
```
binary_only
```
不能仅靠文件名生成
```
asset_card
```
。
先逐文件做 document triage，判断是否应进入资产库，并给出保守的文档角色提示。
只对 triage 通过的文件，从 parsed text 和 file metadata 抽取 canonical
```
asset_card
```
。
先写入
```
upsert_asset_cards
```
，再本地构造保守的
```
merged_assets
```
。
如果存在可确认的归并组，再调用
```
upsert_merged_assets
```
；没有归并组不是失败。

用

complete_pipeline_run

结束 build run，并返回

BuildAssetLibraryData

，必须包含

library_id

、

asset_cards

、

merged_assets

和 summary。

Guardrails

不得虚构 holder、issuer、日期或高置信度。
高歧义字段返回
```
null
```
，不要输出
```
"unknown"
```
。
不确定是否同一材料时，不合并。
不得删除原始
```
asset_card
```
。
这是单人个人资产库；公示、名单、团队材料只有在能保守映射到 owner 时才允许进入。
简历类材料必须视为
```
experience
```
，且不得携带 issuer/date。
没有任何可信材料时，返回结构化
```
partial
```
或
```
failed
```
，不要伪造资产库。
如果
```
document_triage
```
遇到 recoverable model failure，允许 pipeline 保守放行到
```
asset_extraction
```
，但必须留下 warning / audit。