kata-mcp-notion-read

---
name: kata-mcp-notion-read
description: "Query Notion content via MCP. Reading pages, databases, and blocks from Notion via MCP server for local context use"
---

# Kata: Query Notion content via MCP

> **Prefix:** `kata-` | **Type:** Repeatable Skill | **Scope:** Reading pages, databases, and blocks from Notion via MCP server for local context use

## Workflow

```
Progress:
- [ ] 1. Verify MCP preconditions and directives
- [ ] 2. Identify what to fetch
- [ ] 3. Fetch and read the content
- [ ] 4. Present result to user
```

### Step 1: Verify MCP preconditions and directives

1. Consult `.ahrena/.directives` per `lex-directives`.
2. Verify that `notion` is listed in `mcp.servers` (per `lex-mcp`). If not, inform the user and stop.
3. Confirm that the `NOTION_API_KEY` environment variable is defined. If not, inform the user which variable to configure and stop.
4. Consult `codex-mcp-notion` to identify the correct tools and parameters.

### Step 2: Identify what to fetch

1. If the user provided a page ID or URL: use it in Step 3 with `get_page`.
2. If the user provided a database ID: use it in Step 3 with `query_database`.
3. If the user provided a search term: use it in Step 3 with `search`.
4. If no input was provided, ask the user: "Which page, database, or term would you like to look up in Notion?"

### Step 3: Fetch and read the content

**Mode `search`:**
1. Call `search(query="{term}")` to locate matching pages and databases.
2. Present the list of results (title, type, last edited) and confirm with the user which item to expand.
3. For the selected item, call `get_page(page_id="{id}")`.
4. If depth is `full`: call `get_block_children(block_id="{id}")` to retrieve the full content.

**Mode `page`:**
1. Call `get_page(page_id="{id}")` to retrieve metadata and properties.
2. If depth is `full`: call `get_block_children(block_id="{id}")` to retrieve the content blocks.

**Mode `database`:**
1. Call `query_database(database_id="{id}", filter={...})` with optional filters provided by the user.
2. For each returned entry, record: title, relevant properties, page ID.
3. If the user requests details of a specific entry, call `get_page` and `get_block_children` for that entry.

### Step 4: Present result to user

1. Present the retrieved content in a structured, readable format.
2. For databases: display entries as a table with the most relevant properties.
3. For pages: display title, metadata (last edited, creator), and content (summary or full per depth setting).
4. Include the ID and URL of each presented item for future reference.

## Outputs

| Output | Format | Destination |
|--------|--------|-------------|
| Page content | Structured text (title, metadata, blocks) | Response to user |
| Database entries | Table with relevant properties | Response to user |
| Search results | List of matching items with title and type | Response to user |

## Restrictions

- **Read-only:** this kata never creates, modifies, or deletes pages, blocks, or properties in Notion.
- **Use MCP only:** never use the Notion REST API directly; always use MCP server tools (per `lex-mcp`).
- **No hardcoded credentials:** authentication exclusively via `NOTION_API_KEY` environment variable.
- **Confirm before broad searches:** if the query could return many results, present a sample and confirm with the user before proceeding.