Web search
The websearch node searches the web and returns both the model’s answer and the sources themselves, with text excerpts. There is one search node on the platform, and it attaches in two ways: as an ordinary scenario step and as an AI agent tool. The Node mode field in its settings picks which.
Search rides along with a model call, which is why the node has a Model field: the same model searches and writes the answer.
- Execute + Data
- Extension
Two modes
- One search — exactly one search per node run. Works with any model in the list, including free ones and those that cannot call tools, and the cost is known in advance.
- Agentic — the model decides whether to search and how many times. Requires a model that can call tools and must be capped: by the Max searches field and by a spend ceiling the platform sets, which overrides the search counter.
Engine
The Search engine field: Auto (the model’s built-in search when it has one, otherwise Exa), Built into the model, Exa, Parallel, Perplexity. The built-in engine is available only to models that can search on their own — for the rest the node refuses before the call, so you get a clear reason rather than an opaque error from elsewhere.
What the node returns
| Field | What is inside |
|---|---|
| answer | The model's written answer. Empty when “Write an answer” is off. |
| context | A Markdown digest of the sources — the thing to drop into a prompt. |
| results | The sources: index, title, url, content (excerpt). |
| query | The query that was searched. |
| engine / model | The requested engine and the model that ran the request. |
| count / searches | How many sources came back and how many searches ran. |
Turn Write an answer off when you only need the sources: the model writes nothing, and you pay for the search and the inserted excerpts but not for generation. The sources arrive either way.
As an agent tool (a link_extension wire into ai_agent) the node
hands the model exactly that context — the sources themselves, not somebody’s summary.
Tuning
- Max results — how many sources to take (up to 25).
- Excerpt size per source — how much page text to keep.
0leaves it to the engine. In “one search” mode the trimming happens after the response arrives: it caps the payload downstream, not the token cost. - Search depth — the only depth control that reaches a provider’s native search.
- Domains — an allow list and a block list. Some engines accept only one of the two, and the node says so before the call.
Cost
Web search produces two independent charges:
- Tokens — at the chosen model’s ordinary rate. Source excerpts are inserted into the request, so those are real input tokens (five results run about 1,900 of them).
- The search itself — its own line in the ledger. When the price of the search is visible separately from the generation cost, that number is used; when it is not, a flat rate applies. A zero difference is read as “could not separate”, not as “the search was free”.
In “one search” mode the search is charged structurally — even when nothing was found: the search ran and it cost money. In agentic mode the model may not search at all, and then there is no extra line.