Chat With Attachment Response

The Chat With Attachments Response component enables conversational interactions that incorporate file attachments alongside text inputs. This advanced chat component extends beyond standard text-only exchanges by allowing users to upload and reference documents, PDFs, and other files during conversations.

Users can:

• Configure an LLM provider

• Define system prompts

• Specify topics

• Provide context

• Set word limits for both chat history and attachment content

Additional settings allow you to:

• Control attachment size limits

• Define allowed tools for processing different file types

This component is ideal for:

• Document analysis

• Technical support with file sharing

• Content-specific discussions where document-based information is essential

Overview

The Chat With Attachments Response component enhances conversational interfaces by allowing users to include file attachments alongside their text inputs.

This powerful component enables LLMs to analyze, reference, and incorporate information from attached text documents or PDFs while maintaining a natural conversational flow. It's designed for scenarios where users need to share specific files to provide context or ask questions about their content.

---

Key Terms

---

When to Use

---

Component Configuration

---

Required Inputs

---

How It Works

1. File Attachment Processing

   • Uploaded documents are validated against size limits and supported formats.

   • Content is extracted for analysis.

   • Textual content (e.g., PDFs, text files) is prioritized.

   • Summarization is applied to manage word limits.

2. Conversation Flow

   • User input + attachment info are sent to the LLM.

   • System prompt instructs the LLM on how to handle attachments.

   • Responses reference and incorporate attachment data.

   • Chat history + context are maintained across turns.

3. Tool Integration

   • If an MCP Server URL is provided, the LLM can use tools defined in allowed_tools.

   • Tools (e.g., summarizer, analyzer) must exist on the MCP server.

   • Tool results are integrated into LLM responses.

---

Example Use Case: Resume and Cover Letter Review

Scenario:
A user wants AI-assisted feedback on their resume and cover letter before applying for a job.

Configuration:

• LLM Provider: Anthropic Claude-4-Sonnet

• System Prompt:

  "You are a professional career advisor specializing in resume and cover letter optimization.

  Provide constructive feedback focusing on overall impression, content strength, structure, tone, and improvement suggestions."

• Topic: Job Application Review

• Word Limit for Attachments: 2000

• Word Limit for Attachment Summary: 500

• Allowed Tools: ["file_reader", "document_analyzer"]

Conversation Flow Example:

First Interaction: User: "I've attached my resume and cover letter. Could you review them?" Attachments: \[resume.pdf, cover\_letter.txt] AI: Provides detailed resume and cover letter feedback, including improvements. Second Interaction: User: "Thanks! Could you suggest a professional summary for my resume?" Attachments: \[] AI: Suggests a tailored professional summary highlighting skills, experience, and achievements.

---

Best Practices

• Optimize document handling: Set appropriate word limits.

• Craft detailed system prompts: Be explicit about analysis expectations.

• Use clear document formats: Well-formatted PDFs/text files yield better results.

• Manage token usage: Balance comprehensiveness with efficiency.

• Guide users: Inform them about supported formats, size limits, and referencing.

• Test with real documents: Validate performance with diverse samples.

• Enable relevant tools: Ensure MCP server configuration matches required tools.

---

Troubleshooting

---

Limitations and Considerations