PDF.co

PDF.co

· #131 most-used

Parse, build, and deliver any PDF — no human in the loop

DocumentsAutomationFinanceFormsStorage

PDF.co is an API-first document engine covering every stage of the PDF lifecycle: generation from HTML or raw data, OCR conversion of scanned images to searchable text, AI-powered data extraction from invoices and forms, and surgical editing operations like splitting, merging, redacting, and securing files. Connect it to Actionist and your agent can parse incoming invoices without touching a spreadsheet, fill and send contracts the moment a deal closes, split a 200-page batch scan into individual records by barcode, and deliver every output directly to the right tool — all from a plain-English instruction.

Average time saved
11 hours
per person · per month
1 workdays back

Eliminates manual work. Eliminates manual copy-paste between PDF viewers and downstream systems by letting the agent extract, transform, and route document data in one unattended pass.

Schedule

What your PDF.co agent runs on autopilot

A week of scheduled jobs your Actionist agent will execute on your behalf.

28Scheduled jobs
7Agents at work
24/7Always on
Mon
Tue
Wed
Thu
Fri
7am
8am
9am
10am
11am
12pm
1pm
2pm
3pm
4pm
5pm
6pm
Agents
Multi-app workflows

PDF.co × every other app you use

End-to-end automations that span multiple apps — each one a real business outcome.

6Workflows
9Apps spanned
~40 hrsSaved / week
6Personas served
customer-success★ FeaturedSaves 45m saved · runs ~12× /week

Invoice-to-resolution in 60 seconds

When a customer emails a PDF invoice dispute, your agent reads the attachment to extract the invoice number and amounts, then rebuilds a corrected PDF with the accurate line items stamped in — no spreadsheet open, no designer needed. It posts the corrected file to the support Slack channel, books a resolution call on Google Calendar, and the customer receives a reply with the fixed document before your support rep has opened their inbox. Disputes that used to take 45 minutes of back-and-forth close in under a minute.

Trigger: When a new email with a PDF attachment arrives in the support inbox flagged as invoice dispute
Step 1 trigger
Gmail
Watch for incoming invoice dispute email
Step 2 read
Pdfco
Parse invoice PDF to extract amounts and line items
Step 3 write
Pdfco
Generate corrected invoice PDF with accurate figures
Step 4 write
Slack
Post corrected PDF to #customer-success channel
Step 5 write
Google Calendar
Book resolution call with customer
Disputes resolved before the rep opens their inbox
Savings

What this looks like for your team

The comparison strip shows real manual tasks your agent replaces. The calculator translates that into your team's numbers.

Without Actionist
With PDF.co agent
  • Sales
    Proposal re-keying
    Rep copies deal data from the CRM into a Word template, exports to PDF, and emails it manually — 19 minutes per proposal.
    19 min/week
    Sales Agent
    Agent renders proposal PDF
    Agent fills the HTML proposal template with deal values from the CRM and converts it to a polished PDF, delivered to the calendar event.
  • Marketing
    Campaign brief assembly
    Marketer assembles client name, budget, and channel mix into a brief PDF by copying from three different sources each time.
    14 min/week
    Marketing Agent
    Agent builds brief on deal close
    Agent converts the HTML brief template to PDF with client and budget fields populated the moment a HubSpot deal closes.
  • Customer Support
    Invoice dispute re-draft
    Support rep opens the PDF, cross-checks line items in a spreadsheet, recreates the corrected version in a PDF editor, and re-emails it.
    19 min/week
    Customer Support Agent
    Agent parses and rebuilds PDF
    Agent extracts the incorrect figures, generates a corrected invoice PDF, and delivers it to the customer and Slack channel automatically.
  • Human Resources
    Offer letter PDF prep
    HR specialist opens the Word template, types in name and start date, exports to PDF, and sends it — every single hire.
    8 min/week
    Human Resources Agent
    Agent fills and sends offer PDF
    Agent populates the offer letter PDF fields with HR system data and emails the fillable document to the candidate without manual touch.
  • Finance
    Invoice data re-entry
    Finance analyst opens each supplier PDF, types vendor name, invoice number, total, and due date into the accounting system by hand.
    14 min/week
    Finance Agent
    Agent parses and posts invoice
    Agent uses AI Invoice Parser to extract all fields from the PDF and creates the bill record in the accounting system in one pass.
  • Operations
    PO form manual fill
    Ops coordinator copies line items from a supplier PDF into the internal purchase order form across two applications, row by row.
    30 min/week
    Operations Agent
    Agent fills PO from supplier PDF
    Agent parses the supplier's PDF, populates the standard PO form fields, and saves the authoritative record to the document system.
  • Legal
    Redaction before sharing
    Legal associate manually reads through each document, highlights sensitive text, and applies redaction boxes one occurrence at a time.
    6 min/week
    Legal Agent
    Agent redacts by pattern
    Agent runs Search and Delete Found Text to strip all PII or confidential strings matching the defined patterns, producing a clean PDF instantly.

+ 100s of other automations your agent handles

Average monthly savings
11 hours / person
ROI calculator

See what your team gets back

Team size
10 people
Fully-loaded rate
$20 / hour
Hours / week
28
Hours / year
1,400
Annual ROI
$28,000

Baseline: 2.8 hrs saved per person per week, across the full PDF.co automation set.

Connect

How to plug PDF.co into Actionist

Pick the connection method that suits your environment.

The fastest path to PDF.co's full 29-action toolkit. Install the official pdfco MCP server and your agent connects through a permissioned session — no API tokens stored, no credential rotation needed.

1
Open the Apps tab

Find PDF.co in the Apps library and click Connect. MCP is selected by default.

2
Authorise in PDF.co

You'll be redirected to pdf.co to authorise the MCP server access. Sign in with your PDF.co account and confirm the permission grant — the server receives a scoped session credential automatically.

3
Test the connection

Actionist runs a read-only call to verify the handshake. You're ready.

Read the PDF.co docs →
Actions

29 actions your agent can call

Read and write operations available to your Actionist agent.

Triggers

7 events your agent can react to

Events your agent watches for, and the actions it kicks off in response.

Skills

Skills that pair with PDF.co

Reusable agent skills that work well alongside this app.

Pdf

Gives the agent programmatic access to local PDF manipulation — extracting text and tables, creating new PDFs, merging and splitting documents, and filling forms — without an API call, ideal when the file is already on disk.

MCP servers

MCP servers that work with PDF.co

Connect Actionist to MCP servers built for or around this app.

pdfco
Official

The official PDF.co MCP server — exposes the full PDF.co API surface so the agent can extract data, edit, convert, and parse documents with OCR and AI in real time.

AryanBV/pdf-toolkit-mcp
Official

A 16-tool MCP server covering PDF creation, merging, splitting, encryption, form filling, and QR code injection — all via a single connected server.

PDF Generator API
Official

Renders PDFs from HTML or Markdown with full control over page size, margins, headers, and layout — payment handled via x402.

FAQs

Questions about PDF.co + Actionist

What credentials does the agent need to connect to PDF.co?
Via MCP, the agent uses a permissioned OAuth session — no token stored on your side. Via API Token, you paste your PDF.co API key (found under Profile → API Key at pdf.co) into Actionist once; the agent reuses it for every subsequent call. Rotate it in PDF.co if you ever suspect it has been compromised, then update it in Actionist's connected-apps settings.
Which file formats can the agent convert to PDF?
PDF.co's Anything to PDF Converter handles Word (.docx), Excel (.xlsx), PowerPoint (.pptx), images (PNG, JPEG, TIFF, BMP), HTML strings, public URLs, and several other formats. For the reverse direction, PDF to Anything Converter outputs Word, Excel, CSV, PNG, JPEG, XML, and more. If you're unsure whether a specific source format is supported, the Custom API Call action lets you hit the PDF.co endpoint directly and inspect the response.
How does the AI Invoice Parser handle invoices from different suppliers with different layouts?
The AI Invoice Parser uses a layout-agnostic model — it does not require you to define zones or templates per supplier. It returns structured fields (vendor name, invoice number, line items, totals, due date, tax) regardless of where those fields appear on the page. For highly unusual layouts or very poor-quality scans, combine it with the Convert Scanned PDF to Searchable PDF action first to give the parser clean text to work with.
Can the agent process password-protected PDFs?
Yes — use the PDF Security action to remove the password before passing the file to any other action in the pipeline. Supply the owner password (the one that controls permissions) rather than the user password (the one that restricts opening). If you do not have the owner password for an inherited file, PDF.co cannot bypass it — that is by design and is not a limitation of the Actionist integration.
Is there a file size or page-count limit per API call?
PDF.co enforces limits at the account plan level rather than in Actionist. Most plans support files up to 25 MB and several hundred pages per call. For large batches, use the async job pattern: submit via Custom API Call, listen for the Job completed trigger, and retrieve the output URL when it fires. This keeps individual calls well within limits and prevents timeout errors on long-running conversions.
How do I avoid re-processing the same PDF twice in a trigger-based workflow?
The safest pattern is to move or rename the source file after the agent processes it — for example, prefix it with 'PROCESSED_' using Add Text to PDF metadata or move it to a 'done' folder in your storage layer. If you're using the New PDF uploaded trigger, add a condition step that checks the filename prefix before proceeding, so re-uploaded files or retries don't re-enter the pipeline.
Can the agent fill a PDF form if the fields are not interactive (i.e. no AcroForm fields)?
For true fillable PDFs with AcroForm fields, use PDF Filler — it populates named fields directly. For flat PDFs with no interactive fields, use Add Text to PDF to overlay text at the exact coordinates where data should appear, or Add Form Field to PDF to first inject interactive fields and then fill them. For image-only scanned forms, run Convert Scanned PDF to Searchable PDF first so the agent can identify the right zones.
What happens to the output files after the agent processes them — where do they go?
PDF.co returns a temporary download URL for every output file; that URL is valid for a limited window (typically 1–2 hours). Your workflow must pass that URL to a storage step — Google Drive, S3, Notion, or wherever your documents live — before the link expires. The agent does not retain files in PDF.co between runs. Build your workflow to always include a storage write step immediately after any PDF.co action that produces an output file.