skills/notion-query/SKILL.md
Search and retrieve content from a Notion workspace
npx skillsauth add sofer/.agents notion-queryInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Search for and retrieve content from the user's Notion workspace. This skill provides read-only access to Notion via two operations: searching by query and fetching a specific page by ID. It wraps the Notion MCP server tools to give other skills a consistent interface for accessing business documentation.
This skill requires the Notion MCP server to be available. The following MCP tools must be accessible:
mcp__claude_ai_Notion__notion-searchmcp__claude_ai_Notion__notion-fetchIf either tool is unavailable, stop and report: "The Notion MCP server is unavailable. Ensure the Notion MCP integration is configured and running."
Determine which operation to perform based on the input:
mcp__claude_ai_Notion__notion-fetch with the page ID or URL as the id parameter.
https://notion.so/workspace/Page-Title-abc123), Notion Sites URLs (e.g. https://myspace.notion.site/Page-Title-abc123), raw UUIDs (with or without dashes), and data source URLs (e.g. collection://...).mcp__claude_ai_Notion__notion-search with the following parameters:
query: the search query string (required)query_type: set to "internal" for content search (this is the default and most common use)page_url: if a scope was provided and it is a page URL or ID, pass it here to restrict search to that page and its childrendata_source_url: if a scope was provided and it is a data source URL (starts with collection://), pass it here to restrict search to that data sourcefilters: if date range or creator filters were provided, pass them as an object with created_date_range (containing start_date and/or end_date in ISO 8601 format) and/or created_by_user_ids (list of user ID strings)Page ID takes precedence. Follow the fetch mode instructions above and ignore the query.
The output from a search is a list of matching pages as returned by the Notion MCP server. Each result typically includes:
Present the results clearly so the calling skill can identify which page(s) are relevant.
The output from a fetch is the full page content in Notion-flavoured Markdown. This includes headings, text, lists, tables, and any other content blocks on the page. For databases, the output includes the schema and data source information.
Return the content as-is. Do not summarise or truncate. The consuming skill decides what to extract.
tools
Check whether Claude and Codex have equivalent access to shared agent resources, skills, hooks, plugins, MCP servers, permissions, startup behaviour, and provider-specific adapter config. Use when comparing agent environments, debugging missing capabilities after restart, or deciding whether to symlink a resource or configure a runtime.
testing
Record substantive skill use in an append-only local log. Use after choosing or invoking a non-system skill for real work, when a skill is inspected but not used, or when a skill fails to apply. Do not use for routine system skills or incidental file reads.
testing
Turn a vague or underspecified request into a self-contained problem statement. Use when the user has a rough idea, when a request would fail if handed directly to an agent, or before non-trivial work that needs shared understanding.
data-ai
Append a one-line learning to ~/.agents/learning-log.md. Use when the user types /learning, or when something genuinely worth remembering surfaced during work and the user confirms it should be captured.