Searches for messages matching a query.


Method URL:
Preferred HTTP method:GET
Accepted content types:application/x-www-form-urlencoded
Rate limiting:Tier 2
Works with:
Token typeRequired scope(s)

This method returns messages matching a search query.


Argument Example Required Description

Authentication token bearing required scopes.


Search query.

count20Optional, default=20

Number of items to return per page.


Pass a value of true to enable query highlight markers (see below).

page2Optional, default=1

Page number of results to return.

sorttimestampOptional, default=score

Return matches sorted by either score or timestamp.

sort_dirascOptional, default=desc

Change sort direction to ascending (asc) or descending (desc).

Present arguments as parameters in application/x-www-form-urlencoded querystring or POST body. This method does not currently accept application/json.


The matching items are returned as hashes containing contextual messages. This response envelope also contains paging and result information.

Typical success response

    "messages": {
        "matches": [
                "channel": {
                    "id": "C12345678",
                    "is_ext_shared": false,
                    "is_mpim": false,
                    "is_org_shared": false,
                    "is_pending_ext_shared": false,
                    "is_private": false,
                    "is_shared": false,
                    "name": "general",
                    "pending_shared": []
                "iid": "cb64bdaa-c1e8-4631-8a91-0f78080113e9",
                "permalink": "",
                "team": "T12345678",
                "text": "The meaning of life the universe and everything is 42.",
                "ts": "1508284197.000015",
                "type": "message",
                "user": "U2U85N1RV",
                "username": "roach"
                "channel": {
                    "id": "C12345678",
                    "is_ext_shared": false,
                    "is_mpim": false,
                    "is_org_shared": false,
                    "is_pending_ext_shared": false,
                    "is_private": false,
                    "is_shared": false,
                    "name": "random",
                    "pending_shared": []
                "iid": "9a00d3c9-bd2d-45b0-988b-6cff99ae2a90",
                "permalink": "",
                "team": "T12345678",
                "text": "The meaning of life the universe and everything is 101010",
                "ts": "1508795665.000236",
                "type": "message",
                "user": "",
                "username": "robot overlord"
        "pagination": {
            "first": 1,
            "last": 2,
            "page": 1,
            "page_count": 1,
            "per_page": 20,
            "total_count": 2
        "paging": {
            "count": 20,
            "page": 1,
            "pages": 1,
            "total": 2
        "total": 2
    "ok": true,
    "query": "The meaning of life the universe and everything"

Typical error response

    "error": "No query passed",
    "ok": false

Messages are searched primarily inside the message text themselves, with a lower priority on the messages immediately before and after. Search results may be returned in which matches are only on the previous, previous_2, next, or next_2 messages, but not the message itself. When a search query matches more than one message in close proximity to each other only one match will be returned. Using the highlights=true parameter you can identify which items match the query, and which are provided for context only.

If more than one search term is provided, users and channels are also matched at a lower priority. To specifically search within a channel, group, or DM, add in:channel_name, in:group_name, or in:@UserID. To search for messages from a specific speaker, add from:@UserID or from:botname.

For IM results, the type is set to "im" and the property contains the user ID of the target user. For private group results, type is set to "group".

All search methods support the highlight parameter. If specified, the matching query terms will be marked up in the results so that clients may replace them with appropriate highlighting markers (e.g. <span class="highlight"></span>). The UTF-8 markers we use are:

start: "\xEE\x80\x80"; # U+E000 (private-use)
end  : "\xEE\x80\x81"; # U+E001 (private-use)

Please note that the max count value is 100 and the max page value is 100.


This table lists the expected errors that this method could return. However, other errors can be returned in the case where the service is down or other unexpected factors affect processing. Callers should always check the value of the ok params in the response.


No authentication token provided.


Some aspect of authentication cannot be validated. Either the provided token is invalid or the request originates from an IP address disallowed from making the request.


Authentication token is for a deleted user or workspace.


Authentication token is for a deleted user or workspace or the app has been removed.


The workspace token used in this request does not have the permissions necessary to complete the request. Make sure your app is a member of the conversation it's attempting to post a message to.


The workspace is undergoing an enterprise migration and will not be available until migration is complete.


Administrators have suspended the ability to post a message.


The token used is not granted the specific scope permissions required to complete this request.


This method cannot be called by a bot user.


The method was called with invalid arguments.


The method was passed an argument whose name falls outside the bounds of accepted or expected values. This includes very long names and names with non-alphanumeric characters other than _. If you get this error, it is typically an indication that you have made a very malformed API call.


The method was called via a POST request, but the charset specified in the Content-Type header was invalid. Valid charset names are: utf-8 iso-8859-1.


The method was called via a POST request with Content-Type application/x-www-form-urlencoded or multipart/form-data, but the form data was either missing or syntactically invalid.


The method was called via a POST request, but the specified Content-Type was invalid. Valid types are: application/json application/x-www-form-urlencoded multipart/form-data text/plain.


The method was called via a POST request and included a data payload, but the request did not include a Content-Type header.


The workspace associated with your request is currently undergoing migration to an Enterprise Organization. Web API and other platform operations will be intermittently unavailable until the transition is complete.


The method was called via a POST request, but the POST data was either missing or truncated.


The server could not complete your operation(s) without encountering a catastrophic error. It's possible some aspect of the operation succeeded before the error was raised.


This table lists the expected warnings that this method will return. However, other warnings can be returned in the case where the service is experiencing unexpected trouble.


The method was called via a POST request, and recommended practice for the specified Content-Type is to include a charset parameter. However, no charset was present. Specifically, non-form-data content types (e.g. text/plain) are the ones for which charset is recommended.


The method was called via a POST request, and the specified Content-Type is not defined to understand the charset parameter. However, charset was in fact present. Specifically, form-data content types (e.g. multipart/form-data) are the ones for which charset is superfluous.