ScoutOS Changelog

Your best work doesn't all happen in one place. You're in Cursor, Claude Code, or Codex, and Scout is where your agents, automations, and data live. Until now, connecting the two meant custom glue code or a local server to manage.

Not anymore. Scout now has an official, hosted MCP server. Any MCP-compatible client can connect straight to Scout and work with your resources through standard tool calls, so your agents can run a workflow, query a database, or pull a file from Drive without ever leaving the tool you’re already in.

Because it’s hosted by us, there’s nothing to install, run, or keep up to date. And every connection is secure by default, scoped to exactly what your token is allowed to do.

Getting connected

Point your client at the endpoint https://mcp.scoutos.com/mcp and authenticate with your Scout token:

Authorization: Bearer YOUR_SCOUT_TOKEN

Setup takes about a minute in Claude, Claude Code, Codex, Cursor, VS Code, and v0. Here are a couple of examples. In Claude Code, one command adds the server:

claude mcp add --transport http scout https://mcp.scoutos.com/mcp

Or drop this into Cursor, VS Code, or any client that takes JSON config:

{
  "mcpServers": {
    "scout": {
      "url": "https://mcp.scoutos.com/mcp",
      "headers": {
        "Authorization": "Bearer YOUR_SCOUT_TOKEN"
      }
    }
  }
}

Full per-client instructions and verification steps live in the setup guide.

What you can access

Once connected, your client discovers Scout’s tools automatically. From there, you and your agents can:

• Workflows — list, create, and run your Scout workflows
• Agents — list, upsert, and chat with your agents
• Databases — create and inspect collections, tables, and views
• Documents — create, update, and batch-edit table records
• Triggers — list, create, update, and fire triggers
• Drive — upload and download files

Grab a token from your API key settings, point your client at the endpoint, and you’re live.

We can’t wait to see what you and your agents build with it.

You can't use what you can't find. We fixed that.

Most platforms solve for creation. Almost none solve for retrieval. Then navigation quietly becomes the bottleneck.

That's why Universal Search (Cmd+K) is now live in Scout. One keystroke. Everything searchable.

Here's what you'll find with it:

• Universal search — every resource type in your org, from a single surface
• Instant commands — create, navigate, and launch without touching a menu
• Recent-aware — surfaces what you were last working on, no searching required
• Built for scale — build as many agents as you want, and find them…easily

Every second hunting for a resource is a second not using one. Let us get you to work faster.

Log in to Scout and press Cmd+K to check it out.

With Model Management this means you can now:

• Ensure agents only use models your org has vetted and approved
• Keep costs predictable by limiting available options
• Set a cost-efficient default for everyday use while keeping powerful models available for specific agents
• Modify models assigned to existing agents in individually or in bulk

Enhanced Slack Deployments

We simplified the experience to make it easier to view and manage your Scout agents deployed via Slack.

As a reminder, you'll still need to enable the Scout for Slack app to your Slack workspace. In addition, manually add Scout for Slack to any private channels to make them available in Scout.

API Keys

Scout now gives organization admins full control over API credential management. You can create and manage public/private key pairs, attach custom roles and permissions to each one, and enable, disable, or revoke them at any time — all from the console. Permissions follow a simple action:resource pattern, and you can use platform-defined roles or build your own.

The design is intentional: Scout handles authentication and authorization, and you handle everything else — what credentials you issue, who gets them, what they can do, and how long they're valid. Your keys, your rules.

Every publish is now a version. Automation experts can tweak and tune their agents without worrying about breaking things. Need to roll back? One click. Want to see what changed? Version history shows you.

This makes agent development feel like real software development. Iterate safely. Ship with confidence.

Agent Versioning automatically tracks every change to your agents:

• Automatic versioning — Every edit creates a new version
• Version isolation — Test new versions without affecting production
• One-click promotion — Promote to production when ready
• Instant rollback — Revert to previous versions in seconds

Each version is a separate agent ID. The “active” version is simply the one currently designated for production use.

Web Embed for Agents

You can now add a Copilot deployment directly to your agent. Go to agent settings, add a deployment, set it to Copilot, and configure the look and feel. Then copy the HTML code and paste it wherever you want your Copilot to appear.

No more juggling separate Copilot configs. Everything lives in one place.

Examples:

• A marketing-site copilot for product questions
• A support copilot for logged-in customers
• An internal copilot for sales or success teams

Improvements

• Faster Copilot performance
  Copilots now load significantly faster, especially during high traffic, thanks to smarter caching and reduced unnecessary requests.
• Fully isolated preview environments
  Preview deployments are now completely separated from production, making it safer to test changes without any risk to live data.
• Improved performance for large tables
  Large datasets load more quickly, with content now fetched on demand as you interact with it.
• More flexible environment variable management
  You can now more easily retrieve and manage environment variables by filtering for specific names or environments.
• Stronger security protections
  Enhanced safeguards around roles, permissions, and resource ownership help prevent unauthorized access.
• Faster and more efficient secret access
  Reduced latency when accessing secure configuration data.

Last month we shipped agent schedules and file management in Drive. This month, we're adding the piece that ties them together: agent memory.

On its own, memory is a meaningful upgrade. But combined with what we launched in January, it unlocks something bigger: agents that can run on a regular cadence, remember what they did last time, and get better with every run. Let's get into it.

Agent memory

Your agents can now remember things across conversations and across runs.

Before this, every session started from scratch. No context from previous interactions, no awareness of what the agent had already done. That's fine for one-off tasks, but it's a real limitation for anything recurring. An agent prepping you for a Monday standup shouldn't need to relearn your accounts every week. An agent cleaning your CRM shouldn't re-flag records it already reviewed.

With agent memory, agents carry context forward. They know what they did last time, what changed, and what to do differently. Over time, they get more useful because they're building up context rather than resetting it.

You can also continue shaping an agent just by talking to it. Tell it what you liked, what you want adjusted, and it holds onto that going forward.

Better together: memory, schedules, and skills

The real unlock here is how these recent updates layer on top of each other.

Agent schedules (launched in January) let you set an agent to run automatically on a daily, weekly, or monthly cadence. File management in Drive (also January) gives agents a place to create, save, and reference assets across runs. Now add memory, and you have agents that not only run on schedule and produce outputs, but actually learn from each run.

This is also what makes agent skills possible. Skills are reusable capabilities you build into an agent using files in Drive, similar to how coding agents like Claude Code use skills. With memory and files working together, a single agent can handle a much larger variety of tasks reliably and get sharper over time without you having to rebuild it from scratch.

The practical picture: an agent that checks your calendar every morning, preps notes for your calls, remembers what came up in past meetings, and adjusts based on feedback you've given it along the way. Set it up once, and it compounds.

Under the hood

We also did significant platform work this month, including improvements to how agent tables handle embedding models and early groundwork on a more robust API key and permissions system. These will show up as more reliability and flexibility as they roll out over the coming months.