Append Text Lines to a Page
Adds lines of text below whatever is already on a page of an uploaded document, continuing on a new page when the text runs past the bottom.
Overview
Some things have to go onto a document that has no field for them. A contractor's sign-off — the company name and today's date — belongs at the bottom of the page the agency will actually look at, and the bid form it goes on has nowhere structural to put it.
This tool writes lines of text onto a page of a document already in the package. It finds where that page's existing content ends — the lowest text, image or ruled line on it — and starts the block below that, leaving whatever gap the call asks for. Nothing already on the page is moved or erased.
When there is no room left below the content, the lines are not clipped and not shrunk to fit. A new blank page is inserted immediately after the target page and the text continues there. On a page whose content or ruled border already reaches the bottom margin, that happens to the very first line.
What It Does
| Capability | What it means for you |
|---|---|
| Line-by-line placement | Each line is written as its own line, stepping down the page, so numbered lists and blank-line breaks keep their shape |
| Anchored below existing content | The block starts below the lowest text, image or ruled line already on the page — the call sets the gap, not the position |
| Overflow onto a new page | Text with no room left continues on a new blank page inserted right after, instead of being cut off |
| Any page in the document | First page, last page or somewhere in the middle — the target is chosen per call |
| No field required | Text with nowhere to go in the form's structure still gets onto the document |
| Real text | The result is selectable and searchable, not a picture of words |
When the Agent Chooses This Tool
One of the three workflows can call this: bid packages. The appraisal and coating-inspection workflows do not have it.
Even there it is not an open-ended choice. The workflow's instructions name a single use, at a fixed step, once per run: put a three-line contractor sign-off onto the second-to-last page of the uploaded invitation to bid. The page, the horizontal position, the size of the gap and the text size are all written into the instructions. The model supplies the three lines and nothing else.
The third of those lines is today's date, and that is the reason the step exists. The current date is handed to the workflow as context at the start of the run, and the instructions tell the model to use that value and never invent one — typing it here is how a current date gets onto the document. A date that is already printed on the file is a different job: the same workflow corrects the contractor's earlier signature date with Find and Replace Text in a PDF, which changes text that exists rather than adding text that does not.
Adding whole pages that already exist in another document is also a separate tool, Insert Source Pages.
Inputs and Outputs
Takes: the document in the package, which page of it to write on, the lines of text, how far in from the left edge to start them, how big a gap to leave below that page's existing content, and the text size.
Returns: how many lines were written, and how many new pages had to be created. A non-zero page figure is the signal that the text did not fit where it was aimed.
Example
A bid package starts from one uploaded invitation to bid. Early in the run, three lines go onto its second-to-last page: the contractor's name block, and the current date the workflow was handed at the start.
That page carries the form's ruled boxes down to the bottom margin, so there is nothing below the existing content to write into. The three lines land on a new blank page inserted straight after it, and the file grows by one page — which shifts the number of every page after it, including the page-range arithmetic the rest of the run does on that same file.
The lines are real text rather than a picture, so if a name in them turns out to be wrong it can be corrected later with Find and Replace Text in a PDF.
Limits
- This places lines of text; it is not a layout engine. No tables, no columns, no mixed styling within a block, and no flowing around images or existing page furniture. Every line is drawn in one built-in typeface at the size the call gives — it is not matched to the typeface already on the form.
- Lines are placed as given. A long line is not re-wrapped: it runs straight past the right edge of the page, where it stops being visible even though it is still there in the document's text. Text meant to read as a paragraph has to arrive already broken into lines that fit.
- The gap is measured from the page's existing content, not from the top of the page. On a page whose text, images or ruled border already reach the bottom margin, there is no room below them and the whole block goes onto a new page instead of onto the page named.
- Appended text is not a form field. It does not appear when a document's form fields are read, and a form filler cannot edit it.
- The edit replaces the file the package holds. What the package carries afterwards is the edited version, and appending again adds the lines a second time rather than replacing what was written before.
- New pages shift the ones after them. Anything keyed to a page number — a later append aimed at a particular page, an index, a stated total — has to account for the pages this created.
FAQ
What happens if the text doesn't fit on the page?
A new blank page is inserted immediately after the target page, and the remaining lines continue there from the top margin. Nothing is clipped or compressed to make it fit.
Will it write over what's already on the page?
No. The block is anchored to the lowest text, image or ruled line already on the page and starts below it. The call decides how big a gap to leave, not where the block sits. Existing text, fields and artwork are not moved or erased.
Is the appended text searchable, or is it an image?
Real text — selectable, searchable and copyable. That also means it can be corrected later with find-and-replace if a name or date in it turns out to be wrong.
Can I use this instead of filling in the form's fields?
They are different things. What this writes is drawn onto the page rather than stored as a field value, so it does not show up when the document's fields are read and a form filler cannot edit it. Use it for text the form has no field for.
Does it work on a scanned page?
The lines get added, but not on top of the scan. A full-page scan image counts as content reaching the bottom of the page, so there is no room below it and the text goes onto a new blank page inserted after it.
Can text be appended to more than one page?
Each call targets one page, so a document needing text in three places takes three calls. Any new page a call creates shifts the page numbers of everything after it, which matters when the next call names its page by number.