> For the complete documentation index, see [llms.txt](https://developer.celigo.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://developer.celigo.com/mcp/use-cases.md). # What you can do You drive Celigo Platform MCP in plain language. You describe what you want, and the agent picks the tools, chains them, and reports back. You never call a tool by hand. This page collects worked examples: the prompt to ask, what the agent does behind the scenes, and what you get back. They work in any connected client. ## Built-in guidance The server ships more than tools. It also exposes **guided prompts** for multi-step work that your client can invoke directly: account health audits, flow troubleshooting, connection diagnosis, integration planning, flow-config review, and writing Handlebars or SQL. And the **`kb_answer`** tool answers product questions straight from the Celigo Knowledge Base, so the agent can look something up mid-task instead of guessing. ## Examples ### Audit account health > Audit my Celigo account and give me the top issues to fix. The agent lists your integrations and flows, scans recent jobs for failures, checks for offline connections, and returns a prioritized list: which flows are erroring, which connections are down, and where to look first. Tools: `list_integrations`, `list_flows`, `list_jobs`, `list_connections`. ### Diagnose a failing flow > Why did the "Shopify to NetSuite Orders" flow fail last night? The agent finds the flow, pulls its most recent job, reads the per-step results, and inspects the open errors on the failing step. It comes back with the root cause in words ("42 records rejected: NetSuite customer not found"), not raw JSON. Tools: `list_flows`, `list_jobs`, `list_flow_errors`. ### Triage and retry errors > Show me the open errors on that order import, then retry the ones caused by the timeout. The agent lists the step's open errors, groups them by message, and after you confirm, retries the ones you picked. It can also resolve, tag, or assign errors to a teammate. Writes to your destination systems, so the agent confirms first. Tools: `list_flow_errors`, `triage_flow_errors`. ### Build a new integration > Build a flow that syncs new Salesforce accounts into NetSuite customers daily. The agent checks for existing Salesforce and NetSuite connections, fetches the right schemas with `get_schema` so it uses real field names, then creates the export, the import, and the flow, leaving it disabled until you verify it. Tools: `list_connections`, `get_schema`, `upsert_export`, `upsert_import`, `upsert_flow`. ### Inspect or fix a connection > Is my NetSuite connection healthy, and what is it used by? The agent reads the connection (credentials come back masked), reports its status, and traces what depends on it. If you ask it to change a setting, it edits only the field you name and leaves stored credentials untouched. Tools: `list_connections`, `upsert_connection`. ### Run a flow and check results > Run the "Daily inventory sync" flow and tell me what happened. The agent triggers the run, waits for the job, then reports successes and errors per step. Tools: `run_flow`, `list_jobs`. ### Ask a product question > How do delta exports decide what is new since the last run? The agent calls `kb_answer` and returns an answer from Celigo's documentation, with follow-ups in the same thread. Tool: `kb_answer`. ## Tips for better results * **Name the resource.** "the Shopify to NetSuite Orders flow" beats "my flow." The agent matches faster and calls fewer tools. * **Scope to an environment.** Say which account or sandbox you mean if your token can see more than one. * **Let it read the schema before building.** Asking it to "check the schema first" avoids invalid fields and wasted attempts. * **Confirm before writes.** For retries, deletes, and updates, ask the agent to show you the plan first. These actions change live data. * **Work in steps for big jobs.** "List the flows with errors, then we will fix them one at a time" keeps the agent focused and the output readable. --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter: ``` GET https://developer.celigo.com/mcp/use-cases.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.