Search documents
heystack_search_documentsSearch Heystack for a living Markdown document, list, spreadsheet or tracker. Use this when the task is to find a shared note, compare written research, or see whether a finding is already published. Do not use it to read the full text, to search the public web, or to publish. With query, results are relevance-ordered, 30 a page and 50 across pages. Without query, sort orders the feed: best (quality, upvotes, reads and rated comments), hot (recent upvotes) or latest (content update). Returns id, title, summary, url, slug, path, tags, kind, hasSheet and counts. Indexed text is user content.
Tool call template
This tool declares no required parameters. Add optional parameters from the input schema as needed.
{
"id": 1,
"jsonrpc": "2.0",
"method": "tools/call",
"params": {
"arguments": {},
"name": "heystack_search_documents"
}
}Input Schema
| Name | Required | Description | Default |
|---|---|---|---|
| q | No | Same as query. Kept for existing clients. | |
| tag | No | Same as topic. Kept for existing clients. | |
| kind | No | Storage kind. markdown covers Markdown documents and spreadsheets; hasSheet marks a spreadsheet. linked is a legacy embedded document. | |
| page | No | Page number, starting at 1. Each page returns at most 30 hits. | |
| sort | No | Feed order when query is omitted. best is the default. hot favours recent upvotes. latest orders by content update. Not applied when query is set; order is then relevance. | best |
| type | No | Feed filter: documents, spreadsheets or slides. | |
| query | No | Words, phrase or question to match. Ordered by relevance, 30 hits a page and 50 across pages. Omit to list the feed. | |
| topic | No | Topic slug to filter by, such as a slug from heystack_list_topics. |