スリップキーワード検索ノード
Using Slipbox Nodes (Slip Keyword Search)
The Slip Keyword Search node runs a full-text search over the documents in a slipbox. The search term isn't a plain string — it's parsed as a boolean search expression: a space or && means AND (all terms must be present), || or OR means an OR group, and a phrase containing spaces or special characters can be wrapped in quotes to be treated as a single literal string. You can further narrow results with a slip type filter and a tag filter.
This node also works well connected to an agent as a tool, but this tutorial covers it in an Ability to walk through the most basic usage first. Connecting it to an agent will be covered in a separate Slipbox agent integration guide. In this tutorial, you'll build an Ability workflow that runs a keyword search over the documents already in the Agentria Docs slipbox.
Before You Begin
A slipbox must already be set up in your project before starting this tutorial. For instructions on building a slipbox, see the 🔗Slipbox Guide. Since this node is read-only, you'll need a slipbox with a few existing documents (Slips) to search over.
After completing this tutorial, you will be able to:
Use the Slip Keyword Search node to search documents in a slipbox by keyword.
Write search expressions using AND, OR, and quoted-literal syntax as needed.
Narrow your search with a slip type filter and a tag filter.
Verify the structure of the search result (
hits) by running the test and checking the output.
Workflow Overview

The core of this tutorial is the Slip Keyword Search node.
The Slip Keyword Search node takes a search term and runs a full-text search across the whole slipbox (title, summary, and body). Once the node runs successfully, the matching documents are returned as a hits array.
Slip Keyword Search Node Basic Info
The Slip Keyword Search node provides the following options.
Option | Required | Description |
|---|---|---|
| Required | A boolean search expression. A space or |
| Optional | Restricts the search to a specific slip type (Index or Content). Leave blank to search every type. |
| Optional | The maximum number of hits to return (default 20, max 50). If there are more matches than this, the excess isn't returned and |
| Optional | A list of tags every hit must have (AND). Use it to narrow by lifecycle signals like |
| Required | The Slipbox this node searches. Bound at design time in the builder. |
This tutorial connects all five options to the Start Node. Make sure to always fill in Max Results explicitly — see the warning below.
Note: Running this node with
Max Resultsleft blank does not fall back to the documented default of 20 — it was confirmed to return only 1 result instead. Always enter the number of results you want (e.g. 20) explicitly.
Step 1: Create an Ability

Create a new Ability. Enter Slip Keyword Search node for Name and Search slipbox documents by keyword for Description. Which slipbox you search is up to you — this tutorial targets the documents already in the Agentria Docs slipbox.
Step 2: Configure the Start Node

Double-click Start Node to enter the Node Editor. Add the following four variables.
keyword: accepts the search term (a boolean search expression). (String)slipType: accepts the slip type filter. (String)limit: accepts the max results value. (Integer)tags: accepts the tag filter. (Array)
Add tags as an Array type, not a String.
Step 3: Add the Slip Keyword Search Node
Click +Add Node → add a Slip Keyword Search node from the Slipbox category.
Step 4: Configure the Slip Keyword Search Node

Double-click the Slip Keyword Search node to enter the Node Editor. In Target Slipbox, select whichever slipbox you're working with — this tutorial selects Agentria Docs as an example. If you leave this unset, running the node fails, so make sure to check it.
Drag and drop the Start Node's input values onto the remaining fields as follows.
Adapter Variable | Bound Value |
|---|---|
|
|
|
|
|
|
|
|
Once the node runs successfully, it outputs hits (the list of matching documents), slip_ids, total (the number of hits returned), truncated (true if more matches exist), error_code, and error_message.
Step 5: Configure the End Node

Double-click End Node to enter the Node Editor. Add a String output variable named result and drag and drop error_code from the Slip Keyword Search node to bind it. This lets you check whether the search succeeded from the output value.
Step 6: Check What You're Searching

Since this node is read-only, you don't need to create any new Slips. Go to the Agentria Docs slipbox and take note of a few document titles worth searching for (for example, documents with "node" in the title). You'll use a word like this as the keyword input in the next step.
Step 7: Run the Test

Enter a keyword (e.g. 노드, "node") into keyword. For limit, always enter a value explicitly, such as 20 — leaving it blank returns only 1 result. Leave slipType and tags blank, then click RUN TEST. If the result output is SUCCESS, the test succeeded.
Step 8: Check the Debug Log

Check the Debug Log of the Slip Keyword Search node to see the details showing error_code returned as SUCCESS, along with the search results in the hits array (page_id, title, page_type, summary, parent_index_title, tags, snippet). total is the number of hits actually returned for this request, not the total number of matches across the whole slipbox — if there are more matches than limit, truncated is set to true.
Step 9: Cross-Check the Results
Open a few of the document titles returned in hits directly in the Agentria Docs slipbox, and confirm your search term actually appears — not just in the title, but potentially in the summary or body too. Since this node searches the whole document (title, summary, and body), a document without your search term in its title can still show up in the results.
Next Steps
🎉 Congratulations! You've successfully built the "Search Slipbox Documents by Keyword" workflow using Agentria.
Connect this node to an Agent Loop node as a tool, and you can extend it into a workflow where the agent searches slipbox documents on its own during a conversation and uses them in its responses.
Agentria is a place where ideas become reality—your workflow can expand infinitely with your creativity.
For more node examples, visit the 🔗Node Usage page.
Frequently Asked Questions
What does the Slip Keyword Search node do?
The Slip Keyword Search node runs a full-text search over the documents in a slipbox. The title, summary, and body are all searched, and you can narrow results with a boolean search expression (AND/OR/literal) plus a slip type filter and a tag filter. It's a read-only node — results come back as a hits array.
What happens if I enter multiple words in Keyword?
By default, a space is interpreted as AND, so only documents containing all of the words (anywhere in the document, not necessarily together) are matched. For example, searching Slack 노드 matches every document that contains both "Slack" and "노드" ("node") somewhere, even if the two words aren't adjacent. Using || or OR switches to an OR group instead, matching documents that contain either word — a much broader result set.
How do I search for an exact phrase only?
Wrap the search term in quotes. For example, 값 설정 (without quotes) matches any document containing the words "값" and "설정" separately, anywhere — a very broad result. But "값 설정" (quoted, as a literal) narrows the search to only documents where that exact string appears adjacently.
Is it okay to leave Max Results blank?
We recommend against it. The field description states a default of 20, but running it with the value blank (or 0) was confirmed to return only 1 result. Always enter the number of results you want (up to 50) explicitly.
What's the difference between Slip Type Filter and Tag Filter?
Slip Type Filter narrows by a document's structural classification (whether it's an Index document or a Content document), while Tag Filter narrows by tags manually applied to a document (like reviewed or deprecated). Since they're different criteria, you can combine both if needed.