Project Brief // AI Tooling

Inkporter

Vanilla JS HTML CSS Manifest V3
Live Project View Repository
1 MIN_READ

The Challenge // The Problem

AI chat transcripts are locked in platform-specific silos, making it difficult to maintain a local, searchable archive of research, prompts, and co-thinking sessions.

The Solution // Technical Implementation

The Strategy

Inkporter acts as the ‘logistics layer’ for the Inkobytes ecosystem. It utilizes a lightweight, no-dependency approach to DOM scraping, normalizing disparate chat structures from major LLM providers (ChatGPT, Claude, Gemini, Grok) into a unified Markdown schema. Beyond simple export, it provides real-time “thread telemetry”—estimating token usage and conversation tone—to help users manage context density and session quality before archiving.

Core Features

  • Multi-Domain Normalization: Support for major LLM platforms with domain-specific selectors and clean Markdown mapping.
  • Thread Telemetry: Real-time heuristic estimates for Token Usage (against context limits) and Conversation Tone (e.g., “Engaged”).
  • Granular Export Controls: Toggle timestamps, model names, speaker grouping, and YAML frontmatter to suit different archival needs.
  • Flexible Formats: Export as structured Markdown (.md), Plain Text (.txt), or raw JSON for programmatic ingestion.
  • Smart File Management: Automated filename generation based on chat titles and configurable subfolder routing for organized local storage.
  • Privacy-First: Operates entirely locally; no data is sent to external servers or APIs.