Slack integration

Approved channels define the boundary

Retrieve messages and thread replies from the Slack channels an administrator approves. Nothing is read until an administrator approves a specific channel, so the source boundary is a deliberate choice rather than the whole workspace.

Messages, threads, and shared files

Private channels stay with their members

Membership is evaluated when a query runs, not when the message was imported. Someone who leaves a private channel loses access to what was already indexed from it.

Workspace readiness checklist

• A Slack workspace where an administrator can approve the intended channels.
• MemorySync added to each approved channel, which a person must do for private channels.
• A MemorySync project for the imported conversation knowledge.

History window on the first import

The first import reads a bounded window of past conversation rather than the entire archive. Workspaces on a free Slack plan also hide older history at the source, so the window is the smaller of the two limits.

1. 1
   Approve channels

   An administrator selects each channel to index.

2. 2
   Import history

   MemorySync reads the approved channels within the configured history window.

3. 3
   Track new activity

   New messages, edits, and deletions arrive as they happen.

Edits and deletions follow the source

Verify retrieval from an approved channel

Query a memorable phrase from a known message in an approved channel. Confirm the result names the channel and author, and that its link opens the original message in Slack.

Troubleshooting and related sources

Operations on this page

Authentication and scope

Before you start

You need a working Slack connection. Create one with the OAuth flow on Providers & OAuth — Slack cannot be connected with an API key. Everything on this page assumes connection["status"] == "connected".

1. See which channels are available

Reads the channels the installation can see, annotated for a picker, so a user chooses from reality rather than typing a channel name that may not exist or may not be visible to the bot. It takes no parameters and performs no ingestion.

import os
from memorysync import MemorySyncClient

client = MemorySyncClient(
    api_key=os.environ["MEMORYSYNC_API_KEY"],
    base_url="https://api.memorysync.io",
)

available = client.connections.slack.available_channels("conn_8f21a4")
print(available["total"])
for channel in available["channels"]:
    print(channel["id"], channel["name"], channel["selectable"], channel["reason"])

2. Select the channels to sync

Selecting a channel does not import it immediately. It marks it for the next sync, which you trigger on Sync & Jobs.

import os
from memorysync import MemorySyncClient

client = MemorySyncClient(
    api_key=os.environ["MEMORYSYNC_API_KEY"],
    base_url="https://api.memorysync.io",
)

client.connections.slack.add_channels("conn_8f21a4", ["C0123SUPPORT", "C0456PRODUCT"])

3. Read back what is selected

This one returns a bare JSON array rather than an object with a channels key, so iterate the response itself.

import os
from memorysync import MemorySyncClient

client = MemorySyncClient(
    api_key=os.environ["MEMORYSYNC_API_KEY"],
    base_url="https://api.memorysync.io",
)

for channel in client.connections.slack.channels("conn_8f21a4"):
    print(channel["channel_id"], channel["channel_name"], channel["state"], channel["is_active"])

4. Block channels by name pattern

The exclusion policy is a list of channel-name patterns this connection will never sync. It is a standing rule rather than a per-channel decision: a pattern blocks matching channels that are selected now and any that appear later, which is what you want for whole families like incident-* or dm-*.

import os
from memorysync import MemorySyncClient

client = MemorySyncClient(
    api_key=os.environ["MEMORYSYNC_API_KEY"],
    base_url="https://api.memorysync.io",
)

client.connections.slack.set_exclusion_policy(
    "conn_8f21a4",
    patterns=["random", "incident-*", "*-alerts"],
)
policy = client.connections.slack.exclusion_policy("conn_8f21a4")
print(policy["tenant_patterns"], policy["effective_patterns"])

5. Map Slack users to MemorySync users

This is the step that makes Slack memories useful. Without a mapping, a message is not attributed to a person, so querying as that person will not find it. A mapping joins a Slack member id to a numeric MemorySync user id.

import os
from memorysync import MemorySyncClient

client = MemorySyncClient(
    api_key=os.environ["MEMORYSYNC_API_KEY"],
    base_url="https://api.memorysync.io",
)

roster = client.connections.slack.identities("conn_8f21a4")
print(roster["total"], roster["matched"], roster["unmatched"])

client.connections.slack.link_identity(
    "conn_8f21a4",
    slack_user_id="U024BE7LH",
    memorysync_user_id=4821,
)

6. Refresh the roster

Pulls the workspace member list again so newly joined people appear as linkable. It does not create mappings — those stay deliberate.

import os
from memorysync import MemorySyncClient

client = MemorySyncClient(
    api_key=os.environ["MEMORYSYNC_API_KEY"],
    base_url="https://api.memorysync.io",
)

client.connections.slack.sync_identities("conn_8f21a4")

7. Drop a channel

Removes it from the selection so future syncs skip it. Objects and memories already imported are untouched.

import os
from memorysync import MemorySyncClient

client = MemorySyncClient(
    api_key=os.environ["MEMORYSYNC_API_KEY"],
    base_url="https://api.memorysync.io",
)

client.connections.slack.remove_channel("conn_8f21a4", "C0456PRODUCT")

Errors and next action