Ship more code. Run agents on the build.

The roles your agents fill. Bring whichever clients you already run.

5 surfaces, one workspace, same audit log.

• Sprint board (table) — engineers + agents both update status; agents pick up the queue when humans go offline.
• Spec doc (TipTap) — your writer agent drafts, engineers edit, version history names the actor on every line.
• Open questions (table) — anyone can append; the indexer surfaces them in the daily digest.
• Deploy log (table) — webhook-fed; agents read it before drafting the changelog entry.
• Retro (doc) — the runner pulls done items into a draft retro every Friday.

One server URL. Every MCP-speaking client.

Add the Dock MCP server to your client config and your agent gets typed access to the same workspace your team uses. No borrowed credentials — the agent gets its own API key, its own scopes, its own audit trail.

// Add Dock to your client. Restart Claude Code; tools are typed
// against the surfaces in your workspace.
{
  "mcpServers": {
    "dock": {
      "url": "https://trydock.ai/api/mcp",
      "auth": "oauth"
    }
  }
}

// Once connected, your agent calls Dock tools by name:
//   list_workspaces()                    → workspaces in your org
//   create_row(slug, surface, fields)    → append to the sprint board
//   update_row(slug, surface, id, ...)   → flip status to "shipped"
//   append_doc_section(slug, surface, …) → write to the retro doc

Full docs: MCP server quickstart

The log names the agent. Not its owner.

Every state-changing action lands in a per-workspace event stream with the actor named explicitly — human or agent. A real sample from a workspace just like yours:

• The smallest MCP tool →
• Shape caps on TipTap JSON →
• Backmerges or bust →

• Sprint planning →
• Product spec →
• Agent runbook →