Actions

The app provides 29 actions and no triggers, so Nitro is always a step inside a Zap rather than its starting point. Every action takes a file input; Zapier supplies either the file itself or a URL to it, and the app handles both. Action names below are the labels you see in the Zap editor’s action picker.

​

Output Type

Actions that return a file have an Output Type field: Extractions have no Output Type — they always return structured fields. See Working with files for how to use each.

​

Conversions

Notes:

• Input Format on the to PDF conversions is not detected from the uploaded file — pick the format that matches your file or the conversion fails.
• PDF to Image is a multi-file output. See Multi-file outputs.
• PDF to PDF/A conformance levels ending in a (1a, 2a, 3a) require a tagged source PDF; use 2b (the default) or a u level for general archiving.

​

Extractions

All of these return structured fields and have no Output Type. Notes:

• Text Extract with Reading Order on reorders text to follow the visual reading flow. It helps on multi-column layouts but can interleave adjacent columns on dense designs. pages_text is derived by splitting the text on blank lines, so it approximates page boundaries rather than guaranteeing them.
• Text Bounding Box Extract matches literally and case-sensitively unless Treat as Regular Expressions is on. Regex Flags accepts ignore-case, multiline and dot-all.
• Text Bounding Box Extract and Smart PII Detection both emit Redactions JSON, shaped exactly like Redact PDF’s input, so the two chain directly into a detect-then-redact Zap.
• Smart Detect Form Fields outputs map straight into Add Fillable Form Fields — see Chaining actions.
• Parse Invoice handles scanned invoices with built-in OCR.

​

Transformations

All return a file. Notes:

• Compress has no options and always targets the smallest file. Use Optimize to choose the size/quality trade-off.
• Flatten is irreversible — form fields, annotations and layers become static page content.
• Rotate takes two parallel lists. Supply one rotation amount to apply it to every listed page, or as many amounts as page indices to rotate each page differently.
• Split requires a JSON array of arrays: each inner array is the set of pages that becomes one output file. A malformed value fails the step with an InvalidInput error before any request reaches Nitro.
• Merge takes a genuine list, so a Zap can merge up to 25 files in one step — unlike the Power Automate connector, which has eight fixed slots. The merged PDF gets a bookmark per source document (Document 1, Document 2, …); there is no option to turn that off in Zapier.
• Watermark’s Bounding Box is x0,y0,x1,y1 corner coordinates; Redact boxes are [x, y, width, height]. Both are in points from the top-left of the page — see the Bounding Box Guide.
• Password Protect needs at least one of Owner Document Key or User Document Key. Leaving Allowed Permissions empty keeps the document read-only.
• Unprotect requires at least one unlock key and fails with InvalidInput if both are empty.
• Ocr PDF processes one language per request.

​

Forms

Notes:

• Fill PDF Form takes the values either as a file or as inline Field Mappings — one or the other. Checkbox fields accept true, yes, 1 or x. Strict Mode fails the step on unknown field names.
• Add Fillable Form Fields takes one entry per field in each list; map the matching Smart Detect Form Fields outputs into them and the counts line up automatically. The older Form Fields to Create input is legacy and ignored.

​

Worked examples

These three pages document one action each in full — inputs, outputs, and a complete Zap.