Gemini Deep Research → Notion

Execution Mode

Run ALL steps in the MAIN SESSION. Do NOT spawn a subagent.

The browser tool (OpenClaw managed profile) is only available in the main session.
Subagents cannot access the browser, so all browser automation must happen here.

Reply first: "🔬 Deep Research starting for: [topic]. This takes ~25 min. I'll update you when done."

Then execute all phases below sequentially.

Instructions

Complete ALL steps below in the main session.

Phase 1: Trigger Deep Research

1. INLINECODE0
Snapshot, find the text input, type the research query. Always prepend "请用中文回答。" to the query so the research output is in Chinese.
Click "工具" (Tools) button (has page_info icon) → click "Deep Research" in the menu
Click Send to submit the query
Wait for research plan to appear (~10s), then click "Start research" / "开始研究" button

- If snapshot-click doesn't work, use JS:

(() => { var btn = Array.from(document.querySelectorAll('button')).find(b => /Start research|开始研究/.test(b.textContent.trim())); if (btn) { btn.click(); return 'clicked'; } return 'not found'; })()

6. Verify research started: button should be disabled, status shows "Researching X websites..." or "正在研究..."
Save the conversation URL from the browser

Phase 2: Wait for Completion

1. Run exec("sleep 1200") (20 minutes) + INLINECODE4
After waking, check status via JS: INLINECODE5
Look for completion signals: "I've completed your research" or "已完成"
If still running, sleep another 600s and check again (max 2 retries)
If failed/stuck after retries, announce the failure and exit

Phase 3: Extract Report

1. Count message-content elements: INLINECODE6
The research report is in the LAST message-content element (usually index 2)
Get total length: INLINECODE8
Extract in 8000-char chunks using substring: INLINECODE9
Concatenate all chunks into the full report text
Save to a temp file: write full report to INLINECODE10

Phase 4: Export to Notion

Parent page ID: 31a4cfb5-c92b-809f-9d8a-dd451718a017 (Deep Research Database)

1. Read the Notion API key: INLINECODE12
Parse the report into Notion blocks:

- Lines starting with # → heading2/heading3 blocks - Bullet points → bulletedlistitem blocks - Regular text → paragraph blocks - Add a callout at top: "🔬 Generated by Gemini Deep Research on YYYY-MM-DD" - Split rich_text at 2000 chars

3. Create the page via Notion API:

   curl -s -X POST "https://api.notion.com/v1/pages" \
     -H "Authorization: Bearer $NOTION_KEY" \
     -H "Notion-Version: 2025-09-03" \
     -H "Content-Type: application/json" \
     -d '{"parent":{"page_id":"31a4cfb5-c92b-809f-9d8a-dd451718a017"},"icon":{"type":"emoji","emoji":"🔬"},"properties":{"title":{"title":[{"text":{"content":"TOPIC"}}]}},"children":[BLOCKS]}'

4. If >100 blocks, append remaining via PATCH to INLINECODE14
Rate limit: wait 0.5s between batch requests

Phase 5: Announce

Report back with:

- Research topic
Brief summary (2-3 key findings)
Notion page URL: INLINECODE15

Notes

- Always use profile="openclaw" for browser
Deep Research is under "工具" (Tools) menu, NOT the model selector
If Gemini needs login, announce failure — user must log in manually
The full pipeline should complete in ~25-30 min total

Gemini Deep Research → Notion

执行模式

在主会话中执行所有步骤。不要生成子代理。

浏览器工具（OpenClaw 管理配置文件）仅在主会话中可用。
子代理无法访问浏览器，因此所有浏览器自动化操作必须在此处完成。

首先回复：🔬 正在开始深度研究：[主题]。大约需要 25 分钟。完成后我会通知您。

然后按顺序执行以下所有阶段。

说明

在主会话中完成以下所有步骤。

阶段 1：触发深度研究

1. browser action=open profile=openclaw targetUrl=https://gemini.google.com/app
截图，找到文本输入框，输入研究查询。始终在查询前添加请用中文回答。，以便研究输出为中文。
点击工具按钮（带有 page_info 图标）→ 在菜单中点击Deep Research
点击发送提交查询
等待研究计划出现（约 10 秒），然后点击开始研究按钮

- 如果截图点击无效，使用 JS：(() => { var btn = Array.from(document.querySelectorAll(button)).find(b => /Start research|开始研究/.test(b.textContent.trim())); if (btn) { btn.click(); return clicked; } return not found; })()

6. 确认研究已开始：按钮应变为禁用状态，状态显示正在研究 X 个网站...或正在研究...
从浏览器保存对话 URL

阶段 2：等待完成

1. 运行 exec(sleep 1200)（20 分钟）+ process(poll, timeout=1200000)
唤醒后，通过 JS 检查状态：(() => { var el = document.querySelectorAll(message-content)[1]; return el ? el.innerText.substring(0, 200) : NOT_FOUND; })()
寻找完成信号：Ive completed your research或已完成
如果仍在运行，再休眠 600 秒并重新检查（最多重试 2 次）
如果重试后失败/卡住，宣布失败并退出

阶段 3：提取报告

1. 统计 message-content 元素数量：document.querySelectorAll(message-content).length
研究报告位于最后一个 message-content 元素中（通常索引为 2）
获取总长度：document.querySelectorAll(message-content)[2]?.innerText?.length
使用 substring 以 8000 字符为块提取：document.querySelectorAll(message-content)[N]?.innerText?.substring(START, END)
将所有块拼接成完整报告文本
保存到临时文件：将完整报告写入 /tmp/deepresearch.md

阶段 4：导出到 Notion

父页面 ID： 31a4cfb5-c92b-809f-9d8a-dd451718a017（深度研究数据库）

1. 读取 Notion API 密钥：cat ~/.config/notion/api_key
将报告解析为 Notion 块：

- 以 # 开头的行 → heading2/heading3 块 - 项目符号 → bulletedlistitem 块 - 普通文本 → paragraph 块 - 在顶部添加标注：🔬 由 Gemini Deep Research 于 YYYY-MM-DD 生成 - 在 2000 字符处拆分 rich_text

3. 通过 Notion API 创建页面：

bash curl -s -X POST https://api.notion.com/v1/pages \ -H Authorization: Bearer $NOTION_KEY \ -H Notion-Version: 2025-09-03 \ -H Content-Type: application/json \ -d {parent:{page_id:31a4cfb5-c92b-809f-9d8a-dd451718a017},icon:{type:emoji,emoji:🔬},properties:{title:{title:[{text:{content:TOPIC}}]}},children:[BLOCKS]}

4. 如果超过 100 个块，通过 PATCH 将剩余块追加到 /v1/blocks/{page_id}/children
速率限制：批次请求之间等待 0.5 秒

阶段 5：宣布

报告以下内容：

- 研究主题
简要摘要（2-3 个关键发现）
Notion 页面 URL：https://www.notion.so/idwithout_dashes>

注意事项

- 浏览器始终使用 profile=openclaw
Deep Research 位于工具菜单下，而非模型选择器
如果 Gemini 需要登录，宣布失败——用户必须手动登录
整个流程应在约 25-30 分钟内完成

gemini-deep-researchGemini深度研究

gemini-deep-research

Gemini Deep Research → Notion

Execution Mode