Part 3 of 3 · Use

Use when needed

Reuse saved research where the work happens — copy, decks, competitor analysis. Your annotations explain why each capture mattered months later, not just the day you saved it.

What use does

From library to finished work

Use is where compounding pays off. Captures you made months ago inform today's hook or deck, because the note travels with the quote.

Write & decide with evidence

Pull saved hooks, proof blocks, and competitor positioning when you draft a landing page, email, or deck — grounded in real examples, not vague memory.

Annotations that age well

The note you wrote at capture time explains mechanism and intent. That's the difference between a swipe file you use and a folder of screenshots you'll never open.

Export

Export your library as Markdown, CSV, or JSON, for briefs, handoffs, or backup. Take organized research into whatever doc tool your team uses.

Share links

Share a highlight or full annotated page, URL, your marks, and room for someone else to see what you saw. Your annotations travel with the link.

MCP for Cursor & Claude

Connect your library to AI agents. Search, browse projects, and export briefs, so drafts are grounded in what you captured, not generic training data.

Project briefs for agents

Bundle a project's goal, recent captures, and tag themes into context for MCP, or export as Markdown you paste directly into chat.

Saved isn't the same as used

Staring at a blank doc, knowing you saved the perfect reference somewhere
AI drafts that ignore your actual research and sound generic
Sharing a bare URL when your take was the whole point
Launch copy written from scratch when you already collected the patterns
Shipped output

From saved example to draft line

The capture and note from weeks ago, reused in today's landing page hook.

In Gleanit, there are hooks examples in research. I want to come up with a bunch of hooks to play with. It's for the how it works section, so the user will understand what is the flow that Gleanit creates.
Search library Result
Done

Based on what I found in your research collection, here's what we can do.

You've got a solid stash of hooks saved — the Substack "100 high-converting hooks" templates, the Ben Franklin copywork piece from copythat.com, a couple of Reddit posts you flagged for traction, and a definition note on what makes a hook work (surprise, tension, specificity, an open question). Pulling from those patterns, here are some hooks to play with for the how-it-works section, each built to teach the Capture → Find → Use flow:

The contradiction pattern (from your "I failed at everything—and that's why they hired me" note and the "Real Reason Your X Isn't Working" template):

"You don't have a memory problem. You have a retrieval problem."

"The best line you'll write this year, you already read. You just can't find it."

The vivid scene / mid-action pattern:

"You highlight a perfect quote. Three weeks later, you need it. It's gone."

"Open tab. Great line. Close tab. Gone forever. Repeat 40 times a day."

The "how I did X without Y" pattern (straight from your saved Substack template):

"How I write faster without staring at a blank page."

"How to build a swipe file that actually shows up when you need it."

The bold claim pattern:

Use"Storage is easy. Retrieval is the product."

"Your screenshots folder is where good ideas go to die."

The flow + tricolon pattern (your strongest saved ingredient):

Use"Capture with ease. Find when it matters. Use when needed."

This is really useful — having the hooks already collected meant I wasn't generating from nothing, I was remixing patterns you'd personally judged worth saving. That's the entire point of the flow: the research you captured weeks ago becomes the raw material for what you make today, grounded in your taste, not generic suggestions.

You've seen the loop

This is where it compounds

This is where it pays off. The captures you make today become the copy, decks, and drafts you ship months from now.

You've read the whole loop. Now run it on your own research.

Capture a few examples with notes, find them when you write, and ship something grounded in real research.