#Deep multi-source dives

When a question is too big for a single answer, Arachne reads across many real sources and hands you back one structured result you can trust.

#Overview

A deep dive is Arachne's way of handling a large question — one that no single page can answer and no quick reply can do justice. Instead of returning the first thing it finds, Arachne gathers information from multiple real sources, reads them, and organizes what it learns into a structured result you can act on.

This matters because the hardest questions are rarely about finding one fact; they are about pulling many threads together into something coherent. A deep dive does that gathering and organizing for you, and it keeps the sources behind every claim so you can verify any point yourself. As with everything Arachne does, the work is checked before it is reported, so a finished dive reflects a real, verified result rather than an attempt.

#What you can do

With a deep dive inside Arachne, you can:

  • Investigate a broad question that spans many sources rather than one page.
  • Get a synthesized answer, not a pile of links — Arachne reads the sources and pulls them together for you.
  • See the structure, with findings organized into a result you can scan, whether that is a report, a comparison, a table, or another fitting format.
  • Trace every claim through a source pack that gathers the references behind the dive, so you can check the origin of anything Arachne says.
  • Ask for the result in the format you prefer — Arachne chooses the best one automatically, and you can request a different one.
  • Keep the result as an artifact you can reopen, reuse, build on, or export later.

#How to use it

  1. State the question, not the steps. Describe the outcome you want — the thing you are trying to understand or decide — and let Arachne find the path through the sources.
  2. Give it what you already have. Share any links, constraints, scope, or angles you care about up front so the dive starts pointed in the right direction.
  3. Choose a permission level that fits. For a read-and-gather investigation, a cautious level like Look only or Plan first keeps you in the loop while Arachne explores.
  4. Watch the first run. You can follow the work live and steer early if the dive drifts from what you meant.
  5. Read the structured result. Arachne presents what it found in the format that fits — a report, a comparison, a table, or another of the eight — with a source pack alongside.
  6. Refine in plain language. Ask for a deeper look at one angle, a different format, or a tighter focus, and Arachne adjusts.
Tip

Use a quick research ask when you want a fast, synthesized answer to a focused question. Reach for a deep dive when the question is large, spans many sources, and deserves a structured result you will come back to.

#Deep dive vs. a quick research ask

A quick research ask in Arachne gathers from several sources and gives you a synthesized answer fast — ideal when the question is focused and you want an answer now. A deep dive is the bigger sibling: it goes wider, reads more, and organizes what it finds into a structured result. Choose the dive when the breadth of sources and the shape of the answer matter as much as the answer itself.

#Deep dives vs. the Deep Research surface

The deep dive lives inside Arachne, woven into hands-on web work — useful when an investigation is part of a larger task you are already running there. When the whole job is a serious, long-form investigation that deserves a careful, well-organized report, the dedicated Deep Research surface is built for exactly that depth. Both gather their sources so you can verify any point; reach for Deep Research when the report is the point, and for an Arachne deep dive when the investigation supports work already in motion.

Note

Heavier, long-form investigation can draw on credits. Lighter research asks stay simple; the deeper the dive, the more it may use.

#Examples

What you ask:

text
Research how three approaches to remote team scheduling compare, and lay it out side by side.

What you see: Arachne reads across multiple real sources, then presents a comparison putting the three approaches against the things you care about, with a source pack gathering the references behind each point.

What you ask:

text
Do a deep dive on the current landscape for small-business invoicing tools and organize what you find.

What you see: A structured report with sections and findings, drawn from many sources and synthesized into one read, plus a source pack so you can check the origin of any claim — saved as an artifact you can reopen later.

What you ask:

text
Pull together everything notable about this topic from the last year into one table I can sort.

What you see: A table of the gathered findings in rows and columns you can scan and sort, with the sources kept alongside so nothing is unattributed.

#Tips

  • Be specific about the result, not the route. Describe what you want to understand and what a good answer looks like; let Arachne decide which sources to read and in what order.
  • Name your boundaries. Time ranges, regions, the options to include, the angles that matter — constraints make a dive sharper and the result more useful.
  • Lean on the source pack. When a finding surprises you, open the source behind it before you rely on it; a deep dive keeps that trail on purpose.
  • Escalate when the report is the point. If a question grows into something that deserves a careful, standalone report, move it to Deep Research.