Features/AI Bots/Bot File Parsing
descriptionFile Analysis

Bots That Read Your Files.

Drop a PDF, Word doc, Excel spreadsheet, or paste a URL — bots extract the content and answer questions about it in real time.

Supported Formats

picture_as_pdf

PDF

Full text extraction from any PDF — contracts, reports, research papers, manuals.

description

Word (.docx)

Parse Microsoft Word documents, including formatting, tables, and structured content.

table_chart

Excel (.xlsx)

Extract data from spreadsheets — cell values, formulas, multiple sheets, structured data.

language

Web Pages

URL content extraction via Mozilla Readability — articles, documentation, blog posts.

How It Works

1

Upload

Drag a file into the chat or paste a URL. The bot accepts it as part of the conversation message.

2

Extract

FileParser extracts text content — from PDF pages, Word paragraphs, Excel cells, or web article body text via Readability.

3

Respond

The AI reads the extracted content and answers your questions, summarizes key points, or references specific data.

What You Can Ask

summarize

Summarize a 50-page contract

Upload a PDF and ask the bot to extract key terms, deadlines, and obligations.

query_stats

Analyze spreadsheet data

Drop an Excel file and ask “What does this say about Q3 revenue?”

article

Extract key points from an article

Paste a URL and the bot reads the page via Readability, then summarizes the main arguments.

help

Ask questions about any document

“What are the termination conditions in this agreement?” — the bot finds the answer in the file.

Security & Reliability

FileParser includes SSRF protection — URL fetches are validated against private IP ranges to prevent server-side request forgery attacks. Files are parsed in-memory and never persisted to conversation storage.

  • check_circleSSRF-protected URL fetching (private IP validation)
  • check_circleIn-memory parsing — no file content persisted to storage
  • check_circlePer-file MIME type inference from extension
Supported MIME types
// FileParser handles"application/pdf"        → pdf-parse"application/vnd.openxmlformats" → mammoth"application/vnd.ms-excel"   → xlsx"text/html"               → readability"text/plain"              → passthrough

Ready to let bots read your files?

Drop a document in chat and get instant answers.