Skip to content
Rowset Logo Rowset

Start with your first agent dataset

This guide connects a trusted agent to Rowset and creates one small dataset the agent can inspect and update later.

You will use the dashboard for setup, then the agent will use authenticated MCP or REST for the actual dataset work. Public previews stay off unless you explicitly ask to share a read-only browser page.

Use this as the shortest path. After it works, use the broader dataset guide when you need projects, relationships, image columns, exports, or public previews.

Before you start

You need a Rowset account, a trusted agent runtime, and a private place to store an API key such as an environment variable or secret store.

1. Copy the Rowset setup prompt

Sign in to Rowset and copy the dashboard agent setup prompt.

The docs show a masked example:

Set up Rowset for this user.

Rowset MCP URL: https://rowset.lvtd.dev/mcp/
Rowset REST API base: https://rowset.lvtd.dev/api/
Rowset API key: ***
Rowset skill: https://rowset.lvtd.dev/SKILL.md
Rowset skill install: npx skills add LVTD-LLC/rowset

Read the skill URL or install the repo skill before acting. Store the full Rowset API key in a private environment variable named ROWSET_API_KEY; do not print it in logs, screenshots, public chats, generated files, or final responses. Do not commit it, paste it back to chat, or save it in a tracked config file. Configure Rowset as a remote Streamable HTTP MCP server named rowset, using the Rowset MCP URL above and bearer-token env var ROWSET_API_KEY so requests send Authorization: Bearer <key>. For Codex/OpenClaw-compatible clients, the setup command is: codex mcp add rowset --url <Rowset MCP URL> --bearer-token-env-var ROWSET_API_KEY. If the client only supports custom headers, set Authorization to Bearer <key>; use X-API-Key only for REST clients that cannot send bearer tokens. After setup, discover the current MCP tools and API docs at runtime before invoking named tools. Then call get_user_info to verify authentication, get_rowset_capabilities to load the current Rowset feature guide, and get_all_datasets to discover available datasets, get_archived_datasets before restoring archived datasets, and search_datasets when you need filters. If auth fails, confirm the env var contains the full key, not only its prefix. Use search_rows for ranked row search across datasets, or get_dataset before dataset-specific row work so dataset context, schema, and relationships are in context. Use search_dataset_rows for ranked row search within one dataset. Use create_dataset when you need to create a dataset on the fly. Use update_dataset_public_preview when the user asks for a shareable read-only preview.

The dashboard preview masks the API key. The copy button includes the real key, so treat the copied prompt like a password.

2. Store the API key privately

Store the key in the agent runtime as ROWSET_API_KEY.

For MCP and REST, Rowset expects a bearer token:

Authorization: Bearer YOUR_ROWSET_API_KEY

Use X-API-Key only for REST clients that cannot send bearer tokens.

3. Connect the MCP server

Configure the agent's MCP client with the hosted Rowset endpoint:

codex mcp add rowset --url https://rowset.lvtd.dev/mcp/ --bearer-token-env-var ROWSET_API_KEY

The command records the env-var name, not the raw key.

4. Verify access

Ask the agent to verify the connected account and load the current Rowset capabilities:

Call get_user_info, then call get_rowset_capabilities.

The agent should use MCP tool discovery for exact tool names and input schemas.

5. Create one dataset

Ask the agent to create a small dataset with a stable index column. For a first run, choose a workflow with a natural key:

  • email for a personal CRM
  • task_id for an agent task board
  • feedback_id for feedback triage
  • sku for a product catalog

If the source has no stable key, ask the agent to let Rowset generate rowset_id.

Example prompt:

Create a Rowset dataset named agent_tasks with headers task_id, status, owner,
next_action, and notes. Use task_id as the index column. Add three example rows
and include instructions that status must be todo, doing, blocked, or done.

The agent should call get_dataset after creation so it has the dataset key, headers, index column, instructions, and schema context.

6. Try one update

Ask the agent to update a row by index value, then read it back:

Update TASK-001 to status doing, then fetch that row by task_id and summarize
what changed.

You now have a private dataset the agent can continue using in later sessions.

Next steps