Skip to main content
Mintlify

Search documentation

Type to search this documentation.

On this pageOverview

Add the agent to Slack

Install the Mintlify agent in Slack to ask questions, update documentation from team conversations, and capture knowledge in pull requests.

Use the agent in Slack to update your content, ask questions, and capture team knowledge. Mention the agent in a channel to use it collaboratively, or send it a direct message to use it privately.

  1. Navigate to the Automations page in your dashboard.
  2. Click Connect Slack.
  3. Follow the Slack prompts to install the mintlify app in your workspace and grant the requested permissions.
  4. Confirm the connection in your Mintlify dashboard to your Slack workspace.
  5. Test that the agent is working and responds when you:
    • Send a direct message to the agent.
    • Mention the agent with @mintlify in a channel.

To remove the agent from your Slack workspace:

  1. Navigate to the Automations page in your dashboard.
  2. Click Configure Slack.
  3. Click Disconnect to revoke the connection between Mintlify and your workspace.
  4. In Slack, navigate to Manage apps.
  5. Click Remove on the Mintlify app.

When you disconnect the Slack app, the agent stops responding to mentions and direct messages, and automations that post to Slack stop delivering notifications.

To reauthorize the agent, reinstall the app from your Mintlify dashboard. Reauthorize when you need to grant new permissions after a Slack scope update or when you need to recover from a revoked token.

When you reinstall the agent into a workspace that is already connected to your organization, Mintlify refreshes the stored bot token in place. Your existing channels and threads remain intact.

Once connected, you can:

  • Send direct messages to the agent to use it privately.
  • Mention @mintlify in a channel to use it publicly and collaboratively.
  • Attach files and images directly to your messages for the agent to process and include in your docs.
  • Continue conversations in threads to iterate on changes or ask follow-up questions. The agent reuses files, images, and emoji reactions from earlier in the thread as context for follow-up requests, so you don't have to reupload them.
  • Share pull request links with the agent to update related documentation.

Ask the agent questions about your product or documentation and receive answers in the thread.

  • New question: Send a direct message to the agent or mention @mintlify in a channel with your question.
  • Existing thread: Reply in the thread and mention @mintlify with a follow-up question.

Example questions:

  • "Which pages mention the API rate limit?"
  • "How do I set up authentication?"
  • "How is our navigation structured?"

The Slack agent can use connected third-party apps as read-only tools while answering questions or updating your documentation.

  • Shared integrations are available to everyone in your organization.
  • Personal integrations use the account connected by the teammate who is chatting with the agent. For example, a Google Drive search uses that teammate's Drive connection.

You do not have to select integrations for each Slack conversation. Ask the agent to use a connected app naturally in your request, such as:

  • "Compare this Linear issue with the launch plan in the connected Google Drive."
  • "Update the migration guide based on the latest Jira issue."
  • "What does our Notion launch brief say about availability?"

Prompt the agent to connect any integrations you want to use. Complete the authorization link it returns, then send another message in the same thread so the agent can confirm the connection.

See Integrations for supported apps, connection scopes, and permissions.

Use the agent to update your documentation with a new request or in an existing thread.

  • New request: Send a direct message to the agent or mention @mintlify in a channel with instructions on what to update.
  • Existing thread: Reply in the thread and mention @mintlify with instructions on what to update.
  • With attachments: Upload images, diagrams, PDFs, SVGs, code files, or other documents with your message. The agent automatically processes and includes them in your documentation. In a thread, the agent has access to images and files shared in earlier messages, so you can reference them in follow-up requests without reuploading.

The agent reads the context of the request or thread and updates your connected repository. Depending on your review process setting, the agent either opens a pull request or commits directly to your deployment branch.

Ask the agent to target a specific branch if you want changes to land somewhere other than your deploy branch. For example, "open the PR against the staging branch" or "push these commits to docs-refresh." The agent creates the pull request against the branch you name, or pushes commits directly to it when the branch already exists.

When an automation posts a Slack notification about a pull request it opened, reply in that thread and mention @mintlify to continue the work. The agent picks up the full run context, checks out the same branch, and pushes your requested changes as follow-up commits to the same pull request, rather than opening a new one.

You can also share a link to any Mintlify pull request directly with the agent to continue it from a new thread.

When a code change automation opens a documentation pull request, it posts a comment on the triggering code pull request that links to the docs pull request. It then assigns a reviewer by trying the code pull request's author and committers in order until one accepts.

The Slack agent and Automations both use the Mintlify agent. Ask the agent to create, configure, run, or troubleshoot any automation. It can perform every action available to you on the Automations page of your dashboard.

Example prompts:

  • "Create an automation that drafts a changelog entry every time we merge to main."
  • "Set up an automation that updates the docs whenever a Jira issue closes."
  • "Turn off the weekly changelog automation."
  • "Run the code change automation on PR #123."
  • "Review the last month of merges and update the changelog."

You can also ask the agent about automation activity, such as recent runs, failures, or credit use.

  • "How can I make my automations more credit efficient?"
  • "Why did the last run of the code change automation fail?"

You can configure a custom automation to run when someone posts a top-level message in a chosen Slack channel. The agent reads each message, decides whether it warrants a documentation update, and opens a pull request when it does. Set this up when you create or edit the automation on the Automations page.

The mintlify app must be a member of the channel you choose. Invite it with /invite @mintlify before saving the automation.

The agent can read source code from any repository connected through the GitHub App or GitLab integration, even if it isn't your documentation repository. This lets the agent build deeper context to guide documentation updates.

When you mention another repository in your request, the agent clones it on demand and inspects the relevant files before drafting changes.

Example prompts:

  • "Update the authentication guide based on the changes in acme/api-server/src/auth.ts."
  • "Document the new endpoints added in acme/backend#1234."
  • "What repositories do you have access to?"

The agent treats cloned repositories as read-only. It uses them for context, but only writes changes to your documentation repository.

When the agent is making changes to your documentation, it provides live feedback in the Slack thread.

  • Status updates: A live-updating message shows what the agent is doing, such as how many files it has searched, read, and edited.
  • Task progress: When the agent breaks your request into multiple tasks, it displays a checklist so you can track progress on each one.
  • Questions: If the agent needs more information, it presents options for you to select from or reply in the thread with a custom answer.
  • Interruptions: If you send a follow-up message while the agent is still working, it stops the current task and starts on the new one.
  • Completion: When the agent finishes, it posts a summary with a link to the pull request or commit and a link to open the changes in the web editor.

The agent's capabilities in Slack match users' roles.

  • Admins and editors: Can ask questions and make documentation changes. The agent has access to file editing and pull request tools.
  • Viewers: Can ask questions only. The agent always runs in read-only mode for Mintlify tools, regardless of the request.

To change a teammate's permissions, update their role on the Members page of your dashboard.

Chat with @mintlify in the Slack side panel to ask questions or request documentation updates without posting in a channel. The side panel agent behaves like the Slack agent: same permissions, same tools, and the same review process for any changes it makes.

  • Be specific: Tell the agent exactly what you want to know or what you want changed and where.
  • Add context: If a thread doesn't contain all the necessary information, include additional details in your message to the agent.
  • Review carefully: Always review pull requests that the agent creates before merging them.
Suggest an edit

Propose a replacement for this page. The site team reviews it before applying any changes.

Export
Documentation menu