# ae-data-integration

Moves a local file or a Feishu sheet into AE along one pipeline: reads the source's business meaning, drafts a tracking plan for confirmation, turns rows into UE records and uploads.

- skill · `type=skill` · Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, OpenCode · 606 installs · ★ 21
- Author: ThinkingAIAgenticEngine
- License: not stated · policy=link-only · description only; the file stays with the author
- Source: https://github.com/ThinkingAIAgenticEngine/ae-cli/tree/HEAD/skills/ae-data-integration/SKILL.md
- Updated: 2026-10-22
- HTML: https://affpapa.org/en/nevibecode/skills/ae-data-integration

## What it does

The skill runs four submodules in order: Source, Tracking plan, Transform, Sink. It profiles the whole file first and infers what the data means for the business. A tracking plan comes next, and a human has to confirm it before anything is ingested. Only then do rows become UE records headed for the receiver. A file sitting on disk is not a reason to upload it.

Accepted inputs are CSV, TSV, TXT, JSON, JSONL (NDJSON), XLS and XLSX, plus Feishu sheet and Bitable (多维表格) URLs. For those URLs the skill checks the `lark-cli` environment itself — installed, logged in, access granted — and snapshots the document locally: XLSX for sheets, JSONL for Bitable. Tencent Docs, DingTalk, WPS, Google Sheets and Notion have no CLI read path, so you export those yourself.

Local analysis is available too, with nothing sent anywhere. Handing a file to AE Agent is a separate branch and caps at 50 MB per file. Uploads are not idempotent: every `data-integration upload` run submits to the receiver again, so do not re-run it just to look at a response — read the saved JSON from the first run. A batch that times out or drops the network counts as unknown, and resuming it automatically is off the table.

## Install

### Claude Code (native platform)

1. **Skills CLI (Vercel)**
```bash
npx skills add ThinkingAIAgenticEngine/ae-cli --skill ae-data-integration -a claude-code
```

2. **Manually: put the files in place**
- File: `~/.claude/skills/ae-data-integration/SKILL.md` — Copy the whole skill folder; for one project use .claude/skills/ae-data-integration/

### Codex

1. **Skills CLI (Vercel)**
```bash
npx skills add ThinkingAIAgenticEngine/ae-cli --skill ae-data-integration -a codex
```

2. **Manually: put the files in place**
- File: `.agents/skills/ae-data-integration/SKILL.md` — For all projects use ~/.agents/skills/

### Cursor

1. **Skills CLI (Vercel)**
```bash
npx skills add ThinkingAIAgenticEngine/ae-cli --skill ae-data-integration -a cursor
```

2. **Manually: put the files in place**
- File: `.cursor/skills/ae-data-integration/SKILL.md` — Cursor also reads .agents/skills/ and .claude/skills/

### GitHub Copilot

1. **GitHub CLI**
```bash
gh skill install ThinkingAIAgenticEngine/ae-cli ae-data-integration
```

2. **Manually: put the files in place**
- File: `.github/skills/ae-data-integration/SKILL.md`

### Gemini CLI

1. **Skills CLI (Vercel)**
```bash
npx skills add ThinkingAIAgenticEngine/ae-cli --skill ae-data-integration -a gemini-cli
```

2. **Platform CLI**
```bash
gemini skills install https://github.com/ThinkingAIAgenticEngine/ae-cli --path skills/ae-data-integration
```

### Windsurf (Devin Desktop)

**Manually: put the files in place**
- File: `.devin/skills/ae-data-integration/SKILL.md` — The legacy .windsurf/skills/ path still works

### OpenCode

**Skills CLI (Vercel)**
```bash
npx skills add ThinkingAIAgenticEngine/ae-cli --skill ae-data-integration -a opencode
```

### Claude.ai

**Claude.ai**
- Zip the folder and upload it in Settings → Capabilities → Skills

Where a file is given instead of a command, copy the files from the source to that path: https://github.com/ThinkingAIAgenticEngine/ae-cli/tree/HEAD/skills/ae-data-integration/SKILL.md

## When to use it
- load a local CSV or Excel file into AE without hand-building requests
- work out what a file means for the business before ingesting it
- snapshot a Feishu sheet or Bitable into a local file
- turn rows into UE records against a confirmed tracking plan
- inspect a dataset locally without uploading it

## Example prompts
- Import `orders.csv` into AE and show me the tracking plan first
- Analyze `events.jsonl` locally, do not upload anything
- Pull the sheet from this Feishu link and get it ready for AE

## Similar
- [data-analysis](https://affpapa.org/en/nevibecode/skills/bytedance-data-analysis.md) — skill · 3600 installs
- [minimax-xlsx](https://affpapa.org/en/nevibecode/skills/minimax-xlsx.md) — skill · 3800 installs
- [csv-excel-merger](https://affpapa.org/en/nevibecode/skills/csv-excel-merger.md) — skill · 1200 installs
- [elasticsearch-ingest](https://affpapa.org/en/nevibecode/skills/elasticsearch-ingest.md) — skill · 1000 installs
- [Skywork Excel](https://affpapa.org/en/nevibecode/skills/skywork-excel.md) — skill · 361 installs
- [reporting-pipelines](https://affpapa.org/en/nevibecode/skills/reporting-pipelines.md) — skill · 354 installs

From the same set: [thinkingaiagenticengine/ae-cli](https://affpapa.org/en/nevibecode/by/thinkingaiagenticengine/ae-cli.md) — 17 cards

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md