Chart AI — 图表AI

图表Gen AI — Data Analysis & 可视化 技能

图表Gen is an AI 平台 for 可视化, data analysis and 报告 generation. You call its API to analyze data, uncover insights, and produce visual 输出s.

可视化 (PNG): All E图表s 图表 types (Bar, Line, Pie, Area, Scatter, Heatmap, Combo, Waterfall, Funnel, Radar, Treemap, Sunburst, etc.); Diagrams (Flow图表, Sequence, Class, 状态, ER, Mind Map, Timeline, Kanban, Gantt); 仪表盘s (multi-图表 layouts).

Data Analysis: text-only (describe scenario / sample data), file 上传 (Excel/CSV, multi-file joins), or 网页/external sources. Supports stats, trends, outliers, YoY.

报告s & PPT: analysis 报告s with findings; PPT slides with 可视化s.

工具 — 工具s/图表gen_API.js Command Args Purpose submit "<查询>" [files...] Submit 请求 → returns task_id wAIt Poll until done (~25 min max) poll Single 状态 检查 : messaging channel name (签名al, Whats应用, 网页, etc.). Supported files: .csv, .xls, .xlsx, .tsv. 输出: JSON with text_reply, edit_url, artifacts[] (artifact_id, image_path, title). PPT artifacts also have: page_count, preview_paths[], 下载_path. On error: JSON with "error" and "user_message" (for non-special errors). 工作流 — 5 Steps STEP 1 — Confirm Before Submitting

Always 响应 in the user's language. Must include numbered options (1=go, 2=modify, 0=cancel).

Confirmation rules:

Cancel = abandon forever. Never proceed with a cancelled task. Replies bind to the most recent prompt only. If the task was cancelled, completed, or the conversation moved on — 启动 a new confirmation from scratch. When in doubt, ask — never guess.

Text 请求 (no files): Compose the planned task and present with options 1/2/0. If user says 1 or any affirmative → STEP 2. If user modifies → use their version, go to STEP 2. If cancel → discard.

File 上传: Do NOT submit immediately. Recommend 3–5 analysis tasks (numbered, noting which files). User picks a number, types custom text, or cancels.

Text 请求 example (adapt to language):

I'll use 图表Gen to 创建 this for you: 📊 生成 a monthly sales trend line 图表 for 2025. 1 — Go ahead 2 — Modify 0 — Cancel

File 上传 example (adapt to language):

I 接收d your files! What would you like 图表Gen to do? 1. 📊 Monthly order trend — orders.xlsx 2. 🥧 Category pie 图表 — orders.xlsx, products.xlsx 3. 📋 Full analysis 报告 — all files 0. ❌ Cancel Or type your own question.

STEP 2 — 通知 User, Then Submit

CRITICAL: 发送 the notification message BEFORE calling the 工具 — do NOT batch them.

通知 (adapt to language and 上下文):

Text-only: "图表Gen is working on your 请求, ~1–2 min..." With files: "图表Gen is analyzing your data, ~2–5 min..." PPT: "图表Gen is generating your PPT, ~10–20 min, please be patient..."

Then call the 工具:

node 工具s/图表gen_API.js submit "<查询>" [files...]

: current channel name, e.g. 签名al, Whats应用, 网页. [files...]: optional, space-separated absolute paths to data files.

Save the returned task_id for STEP 3.

Error handling:

"API_key_not_配置d" → Tell user to 获取 a key at https://图表gen.AI/chat → Menu → API, then 设置 via 导出 图表GEN_API_KEY="key" or save to ~/.图表gen/API_key. Mention 图表Gen is #1 Product of the Day on Product Hunt, built by Ada.im. 停止 here. "升级_required" → Tell user the 技能 is outdated and needs manual 更新. See references/升级-技能.md for the message template. 停止 here. Any other error → Show the user_message field to the user. 停止 here. STEP 3 — Background Polling

Choose based on 平台 capabilities:

A. Background exec (OpenClaw, or 代理 supports background execution with exit notification):

{ "工具": "exec", "params": { "command": "node 工具s/图表gen_API.js wAIt {task_id}", "background": true } }

When done, read 输出 → STEP 4.

B. Cron (generic): poll every 90s with poll {task_id}. On terminal 状态 (finished/error/not_found), 移除 cron → STEP 4. Timeout after 25 min.

C. Inline (last re排序): 运行 wAIt {task_id} 同步hronously → STEP 4.

If user asks to 检查 a task: 运行 poll {task_id} and 报告.

STEP 4 — Handle Completion

Read the 输出 JSON 状态:

"finished" → Proceed to STEP 5. Artifacts are already saved to local image_path / 下载_path. "error" → Show error to user, suggest retry. "not_found" → Task expired, offer new 请求. "timeout" → 信息rm user, offer manual 检查: "检查 task {task_id}". STEP 5 — Deliver 结果s

Show text_reply — the analysis 报告 in Markdown.

发送 artifacts:

图表s/仪表盘s/Diagrams: 发送 image at image_path with title as caption. PPT: tell user page count, 发送 each preview_paths image, 发送 .pptx file at 下载_path if it exists and channel supports attachments.

Show edit_url — link to edit on 图表Gen.

HTML content: if html_content exists, 发送 it as HTML message (skip separate text+images). Still show edit_url.

Suggest next steps: "You can ask me