Token导航 LogoToken导航TokenDH.com
研究检索敏感数据clawhub未标认证来源可访问clear审计通过

notion-diaryNotion diary 搜索

Agent Skill

用于处理 Notion 页面、数据库、工作区内容和结构化记录。它适合让 Agent 查询知识库、整理页面内容、创建记录或把外部信息同步到 Notion。使用时需要确认集成是否已被授权到目标页面或数据库,并区分读取、追加和覆盖更新;涉及批量写入或修改数据库属性时,应先核对字段名称、属性类型和目标页面。

总安装

4,235

周安装

173

GitHub Stars

公开资料未说明

下载量

1,356
OpenClaw

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

2

许可证

MIT-0

最后核验

2026-05-01

来源状态

来源可访问

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:notion-diary(Notion diary 搜索)
来源仓库:https://github.com/breeze-r/notion-diary
安装命令:
openclaw skills install notion-diary
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 OpenClaw 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

ClawHubOpenClaw
openclaw skills install notion-diary

简介

用中文或英文写日记或简短的 24 小时报告,然后使用用户提供的 NOTION_API_KEY 和捆绑的 Python sc 将它们同步到 Notion 中。

SKILL.md

name
notion-diary
description
Write diary entries or short 24-hour reports in Chinese or English, then sync them into Notion using a user-supplied NOTION_API_KEY and the bundled Python script. Use when the user invokes /notion_diary, asks to write a diary with one of four preset styles, wants command-driven journal input, wants to authorize Notion by filling an API key, wants to sync diary text or photo notes into Notion, or wants a fallback short report generated from the last 24 hours of conversation when no diary was written today.
metadata
{"openclaw":{"emoji":"📝","homepage":"https://docs.openclaw.ai/skills","requires":{"bins":["python3"],"env":["NOTION_API_KEY"]},"primaryEnv":"NOTION_API_KEY","install":[{"id":"brew-python","kind":"brew","formula":"python","bins":["python3"],"label":"Install Python 3 (brew)","os":["darwin"]}]}}
user-invocable
true
homepage
https://docs.openclaw.ai/tools/skills

Notion Diary

Use this skill to turn user input into a diary entry or a short 24-hour report, then sync it into Notion through the bundled Python helper at {baseDir}/scripts/notion_diary_sync.py.

Command Surface

OpenClaw sanitizes the skill name to the slash command /notion_diary.

Accept these command patterns:

  • /notion_diary diary style=<plain|gentle|reflective|lyrical>
  • /notion_diary diary date=<YYYY-MM-DD> style=<...>
  • /notion_diary report24h style=<plain|gentle|reflective|lyrical>
  • /notion_diary auto style=<...>

Chinese aliases are also valid:

  • diary -> 日记
  • report24h -> 日报
  • auto -> 自动

Treat the first token after /notion_diary as the mode. Parse date= and style= when present. Everything else in the message is source material for the entry.

Style Presets

Support exactly four preset styles. Read references/style-presets.md before drafting when the user does not define their own style.

  • plain: concise, factual, diary-as-log.
  • gentle: warm, smooth, personal.
  • reflective: restrained, lived-in, slightly introspective. Prefer this by default.
  • lyrical: more scene and atmosphere, but still grounded.

If the user gives no style, use:

  1. reflective when the material contains emotion, travel, or relationships.
  2. plain when the material is mostly work updates.

Authorization And Runtime Inputs

This skill expects NOTION_API_KEY to be injected by OpenClaw. Read references/openclaw-config.example.jsonc before first use.

Keep this package registry-safe:

  • Do not bundle real API keys.
  • Do not bundle personal page ids, data source ids, or workspace-specific URLs.
  • Keep the example config fully placeholder-based so the skill can be published to ClawHub safely.

Optional environment variables:

  • NOTION_DIARY_DATA_SOURCE_ID: preferred existing target.
  • NOTION_DIARY_DATABASE_ID: existing database id if the data source id is not known.
  • NOTION_DIARY_PARENT_PAGE_ID: parent page used to find or create the Daily Journal database.
  • NOTION_DIARY_DATABASE_NAME: default database title. Defaults to Daily Journal.
  • NOTION_DIARY_DEFAULT_STYLE: default style code. Defaults to reflective.

Prefer NOTION_DIARY_DATA_SOURCE_ID when available. If it is missing, use NOTION_DIARY_DATABASE_ID. If both are missing, use NOTION_DIARY_PARENT_PAGE_ID to locate or create the diary database.

The script creates a date-ready database schema. If you want a visible calendar layout in Notion, create the calendar view once in the Notion UI using the Date property.

Diary Workflow

  1. Parse the mode, date, and style.
  2. Gather the raw diary material from the command body, the attached text, and the current conversation context.
  3. Draft a first-pass diary in the requested style without inventing facts.
  4. After the first pass, ask 1 to 3 targeted follow-up questions that help deepen missing moments, transitions, or emotional shifts.
  5. Revise the diary with the user's answers.
  6. Create or update the Notion entry for that date through the bundled script.

Only skip the follow-up questions when one of these is true:

  • the user explicitly says to sync immediately
  • the material is already complete enough for a polished diary
  • the mode is report24h and the user asked for a short brief rather than a refined diary

Prefer titles in this shape:

  • YYYY-MM-DD | 一句话主题

Prefer a short summary sentence for the Notion Summary property.

Follow-Up Question Rules

Ask questions that improve the diary rather than expand it mechanically.

  • Prefer questions about one missing moment, one emotional turn, or one concrete detail.
  • Ask at most 3 questions.
  • Ask only what would noticeably improve the final diary.
  • Keep the questions short and specific.

Good follow-up examples:

  • 当时你最强烈的感受是什么?
  • 下午去古城的时候,现场的状态更像是轻松、匆忙,还是有点收尾的感觉?
  • 晚上学 AI 时,你是兴奋、平静,还是有点疲惫但充实?

Avoid generic questions like:

  • 还有什么想补充的吗?
  • 今天开心吗?
  • 详细说说。

24-Hour Report Workflow

Use this mode when:

  • the user explicitly invokes report24h or 日报
  • the user invokes auto and there is no usable diary material for today
  • the user asks for a short daily recap instead of a full diary

Before drafting a 24-hour report:

  1. Use sessions_list to identify the current sender's most relevant session from the last 24 hours.
  2. Prefer the literal key main for direct chat when it is clearly the active thread.
  3. Use sessions_history to fetch the recent transcript. Keep includeTools off unless tool outputs are part of the substance the user wants summarized.
  4. Summarize only the last 24 hours of visible conversation for the relevant session. Do not pull unrelated sessions unless the user explicitly asks for a broader recap.

The 24-hour report must stay shorter than a full diary:

  • 1 short summary paragraph
  • 2 to 5 bullets or short paragraphs for key actions, blockers, and progress
  • no embellished storytelling unless the user explicitly asks for it

If a proper diary entry for today already exists and the user asks for auto, prefer updating that diary instead of generating a separate short report.

Photo Handling

When the user provides one or more photos, insert image placement markers into the content before calling the sync script:

  • [[image:1]]
  • [[image:2]]

Each marker must appear on its own line between paragraphs. Pass the real local file paths or public URLs through repeated --image arguments to the script in the same order.

If there are images but no obvious placement, append them after the body.

When image paths or public image URLs are available, the sync should do all three:

  • insert the photos into the page body at the matching marker positions
  • write the photos into the Notion Photos property
  • use the first valid photo as the page cover

Sync Script

Use {baseDir}/scripts/notion_diary_sync.py.

Typical calls:

python3 {baseDir}/scripts/notion_diary_sync.py lookup \
  --date 2026-03-28 \
  --mode diary
python3 {baseDir}/scripts/notion_diary_sync.py sync \
  --date 2026-03-28 \
  --mode diary \
  --style reflective \
  --title "2026-03-28 | 曼谷的最后一天" \
  --summary "曼谷最后一天的收尾、想念和被回应。" \
  --content-file /tmp/diary.txt \
  --image /absolute/path/photo1.jpg
python3 {baseDir}/scripts/notion_diary_sync.py sync \
  --date 2026-03-28 \
  --mode report \
  --style plain \
  --title "2026-03-28 | 24h 简短日报" \
  --summary "过去 24 小时的工作与交流简报。" \
  --content-file /tmp/report.txt

The script will:

  • find or create the diary database
  • create or update the page for the target date and mode
  • store the body in page content
  • keep Title, Date, Mode, Style, and Summary properties in sync
  • best-effort upload local images up to 20 MB through the Notion file upload API
  • populate the Photos property and page cover when photos are provided

Drafting Rules

  • Mirror the user's language unless asked otherwise.
  • Keep facts intact and do not invent details.
  • For plain, optimize for clear sequence and compactness.
  • For gentle, allow warmth but keep it natural.
  • For reflective, use the user's preferred balanced style: first set the day's overall state, then move through concrete events and emotional changes gradually.
  • For reflective, the priority is not “more literary”; the priority is natural pacing, light transitions, and a sense of inner movement without over-writing.
  • Let the sentences breathe a little, but keep them readable and grounded.
  • For lyrical, add atmosphere lightly; do not become purple or melodramatic.

Failure Handling

  • If the Notion API key is missing, explain that the skill needs skills.entries["notion-diary"].apiKey or NOTION_API_KEY.
  • If the integration cannot access the target page or database, ask the user to share the parent page with the integration and provide the correct page or data source id.
  • If image upload fails, keep the diary sync successful and add a short text fallback instead of failing the whole run.
  • If some images fail but others succeed, keep the successful photos in Notion and fall back to text only for the failed ones.
  • If session history is unavailable for report24h, summarize only the visible recent chat and say that the report used a limited context window.

Resources

适合场景

01

OpenClaw 用户查找和安装 Skill 时

02

用户想查找某类 Agent Skill 时

03

需要根据任务场景推荐可安装能力包时

04

需要对比不同来源的安装命令和来源信息时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

补充不同宿主或平台的使用分布数据

能力 5

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

OpenClaw

96.07%
按下载量换算1,303

安全审计

VirusTotal

通过

ClawScan

通过

Static analysis

通过

权限和风险

敏感数据

该 Skill 可能接触密钥、Token、环境变量或敏感配置,应进入高风险复核队列,默认不自动发布。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。当前只有一个来源,正式发布前建议补源仓库或其他目录站核验。

来源信息

继续浏览同类 Skills