notion-lifeos
Skill43 starsChanged 7 months agoMITView on GitHub
User's personal Notion LifeOS (PARA: tasks, notes, projects, Make Time journals) with pre-configured DB IDs + API access — required for any LifeOS interaction. Use for adding tasks, taking notes, journaling, querying todos, searching notes, creating projects, reflection. Triggers: 'add task', 'take a note', 'what do I need to do', 'search my notes', '加个任务', '查笔记', '待办', '记录想法', '反思', '目标偏移'. Not for Notion API docs or workspace admin.
In this repository17 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- bibtex-fetchSkill
- context-auditSkill
- daily-summarySkill
- experiment-registrySkill
- handoffSkill
- init-projectSkill
- paper-analyzerSkill
- paper-self-reviewSkill
- project-reviewSkill
- remote-cluster-agentSkill
- research-leadSkill
- session-teamSkill
- sync-docsSkill
- team-digestSkill
- web-fetcherSkill
The file
We split this skill into the rules it gives an agent
8 rules. About 0 are common, 4 are distinctive, and 4 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈4
Distinctive · estimate
This author's own choices. Worth a look.
4
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 26
Read these as needed — do NOT load them all upfront:
- Very specific: line 67
Database IDs: Read ~/.config/notion-lifeos/CONFIG.private.md for IDs (fallback: CONFIG.private.md in skill dir).
- Very specific: line 69
Make Time dedup: Always check if today's entry exists before creating (scripts/check_today_journal.sh).
- Distinctive: line 72
Post-op confirmation: Always confirm results to the user.
- Very specific: line 84
Use MCP notion-fetch to read the top 15-20 most relevant notes (do NOT fetch all 50 — context budget)
- Very specific: line 94
Use MCP notion-fetch to read those 15 notes
- Distinctive: line 125
Select values change: NEVER hardcode — always fetch schema first
- Distinctive: line 126
Duplicate DB names: Always use databases under the LifeOS root page
≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Notion LifeOS Skill
A Notion life management system based on the PARA method and Make Time framework.
When NOT to Use This Skill
- Notion product questions — pricing, features, API documentation, workspace admin
- Generic productivity advice — "how should I organize my life" without intent to store/retrieve data
- Other note-taking apps — unless user wants to migrate data INTO Notion LifeOS
- Database schema modification — creating new databases, adding/removing properties
Core Principles
- Capture first, organize later — Never let processing block input
- Proactive information surfacing — Use Relations to automatically surface information where it's needed
See JEFF_SU_SUMMARY.md for detailed design philosophy.
Reference Files
Read these as needed — do NOT load them all upfront:
| File | When to read |
| references/schema.md | Before creating/updating any entry (get field names & types) |
| references/mcp-guide.md | When using MCP tools (Claude Code / Claude.ai) |
| references/api-guide.md | When using REST API (OpenClaw / Codex / other agents) |
| references/query-guide.md | When querying/filtering data (tasks by date, done status, etc.) |
| references/advanced.md | For composite intents or error handling |
Operation Guide
- With Notion MCP tools → See references/mcp-guide.md
- With Notion API access → See references/api-guide.md
- For structured queries (filter by date, checkbox, select) → Always use scripts or REST API. MCP
notion-searchis semantic only and CANNOT filter by property values.
Intent Recognition & Database Mapping
| User Intent | Target DB | Method |
| "Take a note about XXX" | Notes | MCP / API |
| "Add a task: XXX" | Task | scripts/create-task.sh / MCP / API |
| "The best thing today was..." | Make Time | MCP / API (check dedup first) |
| "Create project XXX" | Projects | MCP / API |
| "Today's tasks" | Task | scripts/query-tasks.sh --date today |
| "Any unfinished tasks?" | Task | scripts/query-tasks.sh --undone |
| "Today's unfinished tasks" | Task | scripts/query-tasks.sh --date today --undone |
| "Search notes about XX" | Notes | MCP notion-search |
| "Add a resource/reference" | Resources | MCP / API |
| "Query my recent thoughts" | Notes | scripts/query-notes.sh --type Thoughts |
| "List active projects" | Projects | scripts/query-projects.sh |
| "Tasks completed this month" | Task | scripts/query-tasks.sh --done --since YYYY-MM-DD --by-edited |
| "/challenge X" | Notes | scripts/query-notes.sh + CC analysis (see Reflection Commands) |
| "/emerge" | Notes | scripts/query-notes.sh + CC analysis (see Reflection Commands) |
| "/drift" | Projects + Task + Git | scripts/collect-drift-data.sh + CC analysis (see Reflection Commands) |
For Note Type selection and Make Time journal extraction logic, see references/query-guide.md. For composite intents (multiple actions in one message), see references/advanced.md.
Business Rules
- Database IDs: Read
~/.config/notion-lifeos/CONFIG.private.mdfor IDs (fallback:CONFIG.private.mdin skill dir). If missing, search for "LifeOS" root page (NOT "LifeOS Template"). Last resort: guide user to references/setup.md. - Schema-first: Fetch database schema before first write in a session — confirm property names and allowed select values.
- Make Time dedup: Always check if today's entry exists before creating (
scripts/check_today_journal.sh). Update if exists. - Date format: ISO-8601 only (e.g.,
2026-03-08). - Relations: Require target page ID — search for it first.
- Post-op confirmation: Always confirm results to the user.
Reflection Commands
These commands use Notion data + Git history as input, with CC performing the analysis. Output is terminal only — never write to Notion.
/challenge <topic>
Stress-test a current belief against past writing.
- Run
scripts/query-notes.sh --type Thoughts --days 90 --limit 50 - Scan the returned titles for relevance to the topic
- Use MCP
notion-fetchto read the top 15-20 most relevant notes (do NOT fetch all 50 — context budget) - Analyze: identify contradictions, weak assumptions, and evidence gaps
- Output in Chinese: "当前信念 → 矛盾证据 → 更强的替代假设"
/emerge
Surface recurring themes and hidden patterns in recent thinking.
- Run
scripts/query-notes.sh --type Thoughts --days 30 --limit 50 - Scan titles first — select top 15 most interesting/diverse
- Use MCP
notion-fetchto read those 15 notes - Identify: recurring themes, emotional patterns, unasked questions, connections the user hasn't made explicit
- Output in Chinese, grouped by theme, quoting original notes
/drift [N days]
Detect gaps between stated goals and actual activity.
- Run
scripts/collect-drift-data.sh --days N(default 30) - Perform semantic matching (project names like "Thinking with Image" ≠ repo names like "agentic_umm" — match by meaning, not string)
- Analyze:
- Per-project: git commit count + completed task count
- 幽灵项目: active in Notion but zero git/task activity
- 隐性工作: heavy git activity in repos not mapped to any active project
- Time allocation distribution
- Output in Chinese: "项目活跃度矩阵" + "幽灵项目" + "隐性工作" + "建议"
Gotchas (Common Pitfalls)
MCP-Specific
- Checkbox:
__YES__/__NO__, NOTtrue/false - URL property: Prefix with
userDefined:(e.g.,userDefined:URL) - Date property: Split into
date:PropName:start,date:PropName:end,date:PropName:is_datetime - multi_select: Comma-separated string, NOT array
- notion-search: Cannot filter by property values — use scripts/REST API
REST API-Specific
- database_id vs data_source_id: REST API uses
database_id, MCP usesdata_source_id— they differ - Date filter: ISO-8601 only, never natural language
General
- Select values change: NEVER hardcode — always fetch schema first
- Duplicate DB names: Always use databases under the LifeOS root page
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

