Set up your workspace

When you’re done, your workspace has Slack connected for teammate routing and a team agents can ask. Do this before rolling the MCP server out — Halyard is only as useful as the team context behind it.

Sign in

Go to app.usehalyard.ai and sign in with a magic-link one-time code or Google.

Caution

Use your work email. Personal-domain addresses (Gmail, Outlook, and the like) are blocked — Halyard is an organization layer and ties your account to a company domain.

Connect Slack

Slack is how teammates get reached. When an agent asks a question that the knowledge base can’t answer, Halyard delivers it to the right person in Slack and returns their reply to the agent. It is the only live teammate-notification channel today.

In the app, open Settings → Integrations and connect your Slack workspace. This is an org-level OAuth connection that authorizes Halyard’s bot for the workspace.

Note

The whole workspace is eligible — there’s no channel picker, and channel selection is not something you configure here. As a matter of practice, point routing at channels where people already answer product, engineering, customer, and operations questions; a silent integration produces no useful memory.

Add your team

Open Team and add the people who can answer agent questions. Each teammate gets:

• Roles — broad job function, stored lowercase. For example engineer, pm, architect, devops. A teammate can hold more than one.
• Skills — concrete domains, stored Title-Case. For example React, System Design, Product Strategy, Security.

Roles route broadly (“ask an engineer”); skills route precisely (“ask someone who knows System Design”). Give each teammate at least one of each so ask_expert can match them.

Here’s the shape list_team returns once teammates are added, so you can see how roles and skills are stored:

{
  "members": [
    {
      "name": "Priya Sharma",
      "roles": ["pm", "product manager"],
      "skills": ["Product Strategy", "Roadmapping", "Analytics"],
      "availability": "ONLINE"
    },
    {
      "name": "David Kim",
      "roles": ["engineer"],
      "skills": ["React", "TypeScript", "Accessibility"],
      "availability": "ONLINE"
    },
    {
      "name": "James Okonkwo",
      "roles": ["architect", "engineer"],
      "skills": ["System Design", "Cloud Architecture", "Security"],
      "availability": "ONLINE"
    }
  ]
}

Note

Each member carries an availability field (ONLINE, AWAY, or DND). It exists in the data model but isn’t surfaced or editable in the UI yet, so don’t rely on it for routing today.

For how matching actually works, see Your team & routing.

Make summarizing the habit

Connecting Slack turns Halyard into a question router. Capturing answers turns it into memory. Set the expectation, for people and agents both:

• When a human answer changes what an agent does, the agent calls summarize_conversation to file it.
• When an agent finishes meaningful work, it calls summarize_work.

Without capture, the same question gets asked again next week. With it, every answer is searchable the next time — which is the whole point of the layer.

Next

• Source overview — connect GitHub, Linear, ClickUp, and Calendar so work events flow in.
• Your team & routing — how roles, skills, and Slack identity decide who gets asked.
• Make agents use Halyard well — the short rules that get agents searching first and capturing after.