Pull text and tables out of a document

What you get

• The text and the tables of a file your account already holds, named by its artifact id.
• A table comes back as rows, with its header row.
• The two calls below build a Word file, then read it back.

Ask your agent

Connect superbot once.
Other clients are on the install page.

claude mcp add --transport http superbot https://api.superbot.gg/mcp -s user

With no browser, post to the install route instead.

curl -fsS -X POST https://api.superbot.gg/mcp/install -d client=claude-code

Then your agent makes two calls.
The second names the file the first one made.

superbot({"intent": "Build a Word document titled Parts list with one table. Columns: Part, Qty, Unit price. Rows: Bolt 40 0.25, Washer 80 0.05, Bracket 6 3.50."})

superbot({"intent": "Extract the text and tables from artifact <artifact id from the first reply> and show me the table as rows."})

This needs an account.
An agent with none signs you in through /auth.md.
You read back one emailed code.

Two real calls

Both ran on 2026-10-18 from one test account, the second on the file the first built.
Each reply is the agent's own, with its heading cut.

Parts list is built as a Word document with one three-part table.

[artifact (id left out) "Parts-list.docx" application/vnd.openxmlformats-officedocument.wordprocessingml.document 10669]

The table has Part, Qty and Unit price columns with Bolt 40 @ 0.25, Washer 80 @ 0.05 and Bracket 6 @ 3.50.

Parts list (from Parts-list.docx):

| Part | Qty | Unit price |
| --- | --- | --- |
| Bolt | 40 | 0.25 |
| Washer | 80 | 0.05 |
| Bracket | 6 | 3.50 |

The only text besides that table is the heading, "Parts list".

Limits

Files

A file your account already holds, named by its artifact id. The calls here read a Word file.

Needs

A superbot account.

Related

• Build a PDF, spreadsheet or deck from a spec
• Read a web page as clean text
• All use cases