PostalForm’s developer and agent surfaces have been moving beyond a single “upload a PDF” path.
PDF upload is still important, but agents and software systems often start with different inputs: plain text, HTML, Markdown, RTF, DOCX, images, form payloads, or bulk templates with merge fields. The product now documents those paths more explicitly and exposes more machine-readable discovery for agents and API crawlers.
What changed
Recent public docs now describe support for:
- Uploading PDFs, DOCX, HTML, Markdown, RTF, image, and plain-text files for conversion, preview, printing, and mailing.
- Agent-created letters from raw text, HTML, Markdown, or RTF.
- Server-rendered PDF previews before payment where available.
- Typed or drawn signatures for supported letters and forms.
- Guided workflow forms and packet workflows.
- Bulk mailings with recipient lists and merge templates.
- Public discovery surfaces for AI assistants, MCP clients, OpenAPI clients, and x402-aware runtimes.
Why discovery matters
Agents do not browse products the same way people do. They need structured instructions, manifests, API descriptions, pricing hints, safety boundaries, and a clear answer to when a human should stay in the loop.
That is why PostalForm publishes pages such as /ai, /agents, /developers, /openapi.json, MCP discovery documents, and x402 discovery. The goal is to make it easier for software to understand when PostalForm is relevant, and harder for it to misuse a physical-mail action.
Source note
This post is based on PostalForm’s public AI instructions, Agents Guide, and Developer Docs.