> ## Documentation Index > Fetch the complete documentation index at: https://www.landbase.com/docs/llms.txt > Use this file to discover all available pages before exploring further. # MCP prompt cookbook > Copy and adapt safe prompts for common Landbase MCP workflows. Use these prompts to tell an MCP client what you want Landbase to do. The client should select the tools, preserve the returned identifiers, and explain any approval or credit requirement before starting paid work. ## Rules for agents using Landbase MCP * Call `list_accounts` when account context is ambiguous. * Preserve the selected `account_id` on every account-scoped call. * Use fields returned by `get_dataset_fields` instead of assuming a schema. * Poll returned handles and run IDs before retrying a request. * Preview and quote paid work before asking for confirmation. * Never claim a dataset or workflow completed without checking its terminal status. ## Choose an account safely ### Prompt ```text theme={null} Use Landbase to list my accounts. If there is more than one, ask me which account to use before doing anything else. Keep the selected account_id for every follow-up call. ``` ### Expected behavior The assistant calls `list_accounts`, asks the user to choose when necessary, and passes the selected `account_id` on later account-scoped calls. `list_accounts` itself does not require an account ID. ### Cost and safety This is an inspection call. It should not spend credits or change data. ## Search companies without spending credits ### Prompt ```text theme={null} Find 20 B2B SaaS companies in Texas with 50–200 employees. Show the search interpretation, row count, available fields, and a preview. Do not enrich, qualify, research, publish, or download anything yet. ``` ### Expected behavior The assistant uses `search` in default mode, keeps the returned session and search handle, and polls `get_search_result` if the search is still running. It should inspect the resulting dataset before proposing paid work. ### Cost and safety Searching and previewing should remain read-only. The assistant must not start enrichment or another paid workflow from this prompt. ## Search people using returned fields ### Prompt ```text theme={null} Using the selected Landbase account, find Marketing VPs at the companies in the current dataset. First inspect the dataset fields and use the actual company and title fields returned by Landbase. Preview the matches and do not reveal contact details yet. ``` ### Expected behavior The assistant calls `get_dataset_fields` before constructing the follow-up request, uses the returned field names, and previews the result before contact enrichment. ### Cost and safety Do not assume that every dataset has the same schema. Contact details and enrichment may spend credits and require a separate approval. ## Match one company or person ### Prompt ```text theme={null} Match Acme's company record in Landbase using its website https://acme.example. Return the match confidence and the identifiers, but do not enrich or update anything. ``` For a person: ```text theme={null} Match Ada Lovelace at Analytical Engines using the strongest identifiers available. Treat a plausible match as unverified and show me the returned record before doing anything else. ``` ### Expected behavior The assistant uses `match_company` or `match_person`, prefers a website, LinkedIn URL, email, or company context over an ambiguous name, and explains whether the match is verified or only plausible. ### Cost and safety Matching is an inspection step. Do not silently chain it into enrichment or contact discovery. ## Limit a dataset before paid work ### Prompt ```text theme={null} Show me the row count and fields for the current dataset. Then create a separate dataset containing at most 50 rows with a read-only transform. Do not enrich or qualify the original dataset. ``` ### Expected behavior The assistant calls `get_dataset`, `get_dataset_fields`, and `transform_dataset` with a read-only `SELECT`, polls the transform workflow, and uses its `output_dataset_id` for later work. It keeps the original dataset separate. ### Cost and safety Transforming a dataset helps control the rows processed by later paid workflows. It does not rank rows unless the SQL explicitly sorts them. ## Enrich after approval ### Prompt ```text theme={null} For the limited dataset, request company employee count and industry enrichment. First show the credit quote or row count and wait for my approval. Do not start the enrichment until I approve it. ``` ### Expected behavior The assistant calls `enrich_dataset` without `confirm`, shows the returned estimate or row count, repeats the same call with `confirm: true` only after approval, and polls the returned workflow. It uses the workflow's `output_dataset_id` after success. ### Cost and safety Enrichment spends credits. Preserve the selected `account_id` on the quote, confirmation, and every workflow poll. ## Qualify or research with explicit context ### Prompt ```text theme={null} For the limited company dataset, ask whether each company sells software to sales teams. Use only the actual company name and website fields returned by get_dataset_fields. Show the quote or row count and wait for approval before starting. ``` For research: ```text theme={null} For each company in the limited dataset, research its public pricing model in one sentence using the website field. Show the proposed research task and credit quote first. Do not begin until I approve. ``` ### Expected behavior The assistant uses `qualify_dataset` or `research_dataset`, supplies the relevant returned context columns, follows the quote/confirmation flow, and polls the resulting workflow. ### Cost and safety Qualification and research can spend credits. Research-mode `search` uses a confirmation turn in the original session rather than a `confirm` boolean. ## Recover a running operation ### Prompt ```text theme={null} Resume the existing Landbase operation from the handle or workflow ID already in this conversation. Check its current status before starting anything new. Keep the same account_id and use the finished output_dataset_id when it succeeds. ``` ### Expected behavior Use `get_search_result` for a search handle, `get_agent_run` or `wait_for_agent_run` for an agent run, and `get_workflow_run` for a workflow run. A wait expiration means the remote work may still be running. ### Cost and safety A timeout or lost response does not prove that work stopped. Never repeat a paid start, upload, publish, or download call until the original operation has been recovered or its outcome is known. ## Publish and download after review ### Prompt ```text theme={null} Export the current working dataset as CSV. Tell me the row count and whether publishing or downloading will spend credits. Ask for approval before creating the file or returning a download URL. ``` ### Expected behavior The assistant publishes when required, polls the publish workflow, uses its `output_dataset_id`, requests a download URL without confirmation first, and repeats with `confirm: true` only after approval. ### Cost and safety Download links are temporary. Download the file promptly and treat the URL as a credential while it is valid. ## Related guides * [Connect to Landbase MCP](https://www.landbase.com/docs/guides/mcp/get-started-with-mcp.md) * [Connect Landbase to Grok Bot](https://www.landbase.com/docs/guides/mcp/grok-bot.md) * [Build your first prospect list](https://www.landbase.com/docs/guides/mcp/first-prospect-list.md) * [Supported tools](https://www.landbase.com/docs/guides/mcp/mcp-supported-tools.md) * [Limits and credits](https://www.landbase.com/docs/guides/mcp/limits-and-credits.md) * [Continue a search or workflow](https://www.landbase.com/docs/guides/mcp/continue-async-work.md) * [Troubleshoot MCP](https://www.landbase.com/docs/guides/mcp/troubleshooting.md) This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.