Create or update Notion pages that readers can find and understand later. Check for existing content, choose the right location, organize conclusions with sources, and preserve relevant material when editing an existing page.
Write a page into Notion that someone else can find and trust — pick the right parent, check for the page that already exists, and structure it for a reader who was not in this conversation. Use when the user says to write up, document, save, or capture something in Notion, or to update an existing page.
Run score Not tested
See how to installThis skill organizes conversation conclusions and other information worth keeping into Notion pages that people can read later. It searches for related pages first and prefers updating an existing page when the topic matches. It states the page’s purpose at the start, arranges the content in the order readers will use it, and includes relevant dates, people, and source links.
Use it to turn a meeting into shared documentation or update existing instructions and explanations. It can choose a team or database where the intended readers can find the material. For database pages, it also fills in properties used for search and filtering, helping people who were not part of the conversation understand the document.
A Notion tool connection is required. Before writing substantial content, the skill presents the title, parent page, and outline for review. It asks where to save the page if the location is unclear and preserves content outside the requested changes. It adds to dated records such as meeting notes and incident reports instead of rewriting them. After writing, it returns the page URL.
This overview was written by AI (OpenAI Codex) from the public SKILL.md. Check the original below for the exact behavior.
AI checked the SKILL.md for operations that need care and for setup requirements. The score comes only from actually running the skill (hands-on test).
Not counted in the score. Rated by AI (OpenAI Codex) as of Oct 29, 2026. Treat it as a guide and read the original before installing.
Review the instructions the agent will receive before you install.
An unfindable page is worse than no page: it splits the truth in two and neither copy gets maintained. Most of the work happens before you create anything.
Search before you create, using the words a previous author would have used, not the words in your prompt. Finding it changes the task from "write a page" to "update a page" — which is almost always the better outcome, because the existing page already has the links, the history, and the readers.
Update in place when the new content belongs to the same subject. Create a new page only when the subject is genuinely new, or when the existing page is a record of a moment — a meeting note, an incident write-up — that should not be rewritten after the fact.
A page's parent determines who ever sees it. Ask which team or database it belongs to rather than defaulting to the first workspace root you can write to. When a database is the right home, fill the properties — an entry with empty properties falls out of every filtered view its readers use.
If you cannot tell where it belongs, ask. A page in the wrong place is harder to find than a page that does not exist, because it looks like coverage.
Open with what the page is for, in one sentence, before any background. Then the content, in the order it gets used rather than the order you discovered it. Date anything time-bound and name the people involved — "we decided" is unreadable in six months.
Link the sources: the issue, the PR, the page you took numbers from.
For anything substantial, show the user the title, the parent, and the outline before writing. Parent and title are what everyone else navigates by, and both are awkward to correct once links exist.
After writing, hand back the URL.
When updating, read the current content first and preserve what you are not changing. Append to a page with history rather than replacing it, and keep the existing headings unless they are wrong — other pages and people link to them.
No reviews yet. Tell others what you thought.
No login needed. Share how it worked for you, ask a question, or note where it did not run.
No comments yet. Be the first to write one.
Review how you use Superset and identify unused features that fit your workflow. Get a short list of suggestions, learn about each option, and set up selected features such as scheduled agents, parallel workspaces, or shared tasks.
Set up recurring tasks in Superset with an agent prompt, a schedule or event trigger, and a chosen execution workspace. After explicit confirmation, create the automation, check its saved settings, and review a real first run.
Browse pages, click controls, fill fields, and inspect screenshots or console messages through Superset's browser pane. Use it to preview a running web app, check interface behavior, and work through browser tasks in a visible session.
Build Stripe payments, subscriptions, and invoicing using verified documentation and API parameter definitions. The skill guides product selection, sandbox testing, and webhook handling, where server notifications determine payment and subscription state.
Verify a UI change or reproduce a bug in the running desktop app using the Chrome DevTools Protocol (CDP), a browser debugging interface. Follow the real user journey and compare screenshots with measurements such as route, scroll position, and persisted state.
Diagnose failed GitHub Actions runs by finding the first meaningful error, separating code problems from infrastructure issues and intermittent failures, and comparing the base branch. Get a focused report explaining the PR’s involvement and recommended next steps.