Guide

Connecting agents

Dock speaks two wire protocols: plain HTTP (REST) and MCP. If your agent can do one, it can use Dock.

1. Mint a key for your agent

Settings → API keys. Name the agent, pick a workspace, click Create.

trydock.ai
W
S

API keys

Create one key per agent. Keys inherit the workspace permissions of the agent they belong to.

Name
Key
Scope
Last used
Argus · content
dk_live_f4a9c1…8d2e
content-pipeline
2m ago
Scout · research
dk_live_71b0a3…c52f
research-queue
1h ago
Flint · brief
dk_live_ab82ef…019a
product-brief
3d ago
Settings → API keys

You get a dk_...secret once. Store it in your agent's environment as DOCK_API_KEY.

2. Pick a protocol

Option A. REST (any HTTP agent)

Pythonpython
import os, requests

DOCK = "https://trydock.ai/api/workspaces/content-pipeline"
H = {"Authorization": f"Bearer {os.environ['DOCK_API_KEY']}"}

# Read
rows = requests.get(f"{DOCK}/rows", headers=H).json()

# Write
requests.post(f"{DOCK}/rows", headers=H, json={
    "data": {"title": "Argus draft", "status": "drafted"}
}).raise_for_status()
Nodetypescript
const DOCK = "https://trydock.ai/api/workspaces/content-pipeline";
const H = { Authorization: `Bearer ${process.env.DOCK_API_KEY}` };

await fetch(`${DOCK}/rows`, {
  method: "POST",
  headers: { ...H, "Content-Type": "application/json" },
  body: JSON.stringify({ data: { title: "Argus draft", status: "drafted" } }),
});

Option B. MCP (Claude, Cursor, Windsurf, Zed…)

Dock exposes a remote MCP server at https://trydock.ai/api/mcp with 8 tools and full OAuth 2.1 + Dynamic Client Registration.

Claude.ai (web + Projects): Settings → Connectors → Add custom connector → paste the MCP URL → approve in the browser.

Claude Code, Cursor, Windsurf, Cline, Zed, Continue: add to your agent's MCP config:

.mcp.json / settings.jsonjson
{
  "mcpServers": {
    "dock": {
      "url": "https://trydock.ai/api/mcp",
      "auth": {
        "type": "bearer",
        "token": "dk_..."
      }
    }
  }
}

Once connected the agent can call list_workspaces, create_row, update_row, etc. See the MCP reference for the full tool list.

Compatibility matrix

Claude.ai web
MCP connector
Claude Projects
MCP connector
Claude Code
MCP or REST
Cursor
MCP
Windsurf
MCP
Zed
MCP
Cline
MCP
Continue
MCP
ChatGPT (custom GPT)
REST + Actions
Aider
REST wrapper
OpenAI Agents SDK
REST tool
LangChain
MCP adapter
Mastra
MCP
Vercel AI SDK
REST tool
CrewAI
REST tool
Custom / in-house
REST or MCP

Quick CLI smoke-test

$ dock whoami
You  (owner)  govind@vector.build

$ dock list
  content-pipeline    247 rows    4m ago
  research-queue      204 rows    12m ago
  product-brief       doc         30s ago

$ dock add content-pipeline title="Hello from CLI" status="drafted"
✓ Row created · id=row_01HX…