TCG Haven Portfolio

The description states:

• TCG Haven Portfolio is a local Python and Streamlit application.
• It uses SQLite, SQLAlchemy, and Alembic migrations for data handling.
• It imports purchase data, reviews card identities, confirms costs, and finalizes orders through four stages.
• It supports OCR-based market observations, with plausibility warnings and manual confirmation steps.
• It handles shipping and transaction fee allocation across inventory lots.
• The system is designed to block unsafe inventory changes until human review is complete.

Inference:

• The product is a local desktop or browser-based tool, not cloud-hosted.
• It is not a SaaS offering; it's a self-contained application for personal use or small business internal workflows.

The description states:

• The author runs TCG Haven, a small trading-card business in Germany.
• The tool was built to replace a fragmented spreadsheet workflow.
• It is positioned as a safer local application for managing card inventories and market tracking.

Inference:

• The product is not marketed externally, but rather developed for internal use and then shared publicly.
• It does not claim to be a general-purpose marketplace or inventory platform — it’s tailored for TCGs with complex variant structures.

The description states:

• The tool is intended for small TCG businesses like the author's own.
• It addresses inventory and market tracking challenges in TCGs, where card variants (set, language, finish) can be difficult to track manually.

Inference:

• The ICP is likely small-scale TCG collectors or retailers, not large distributors or online marketplaces.
• There is no evidence of segmentation beyond this niche use case.

The description states:

• No pricing model, revenue, or monetization strategy is described.
• The tool is a local application and not sold or offered as a service.
• It was built for internal use and then submitted to a hackathon.

Inference:

• There is no evidence of a commercial business model.
• No pricing information, subscriptions, or licensing details are provided.

The description states:

• Built with Python, Streamlit, SQLite, SQLAlchemy, and Alembic.
• Uses GPT-5.6 via Codex for development assistance (not runtime integration).
• Includes 174 automated tests, Playwright browser scenarios, and OCR safety checks.
• The demo is reproducible and anonymized, with synthetic data.
• Privacy auditing was performed.

Inference:

• The tool is developer-focused, not user-facing in a polished way.
• It uses modern Python tools for local development and testing, but no evidence of production deployment or scalability.

The description states:

• The project started as an internal prototype.
• It was substantially extended during Build Week.
• A public demo repository exists with synthetic data.
• No real-world usage, customers, or adoption is reported.

Inference:

• There is no evidence of traction, revenue, or customer base.
• The tool has not been deployed in a production environment beyond the demo.

The description states:

• No mention of competitors or market positioning against other tools.
• It is designed for small TCG businesses, not general inventory systems.

Inference:

• The competitive landscape is not described.
• There is no evidence of existing tools in this niche, nor any comparison to them.

The description states:

• The tool is local-only, not cloud-hosted or scalable.
• It uses GPT-5.6 for development only, not runtime AI.
• No real-world usage or adoption is reported.

Inference:

• Risk of limited scalability due to local nature and lack of cloud infrastructure.
• Risk of no commercial viability without a clear path to monetization or customer base.
• The tool may be too niche or experimental for broader market appeal.

1. Has TCG Haven (the business) actually adopted this tool in production?
2. What is the actual size and scope of the TCG Haven business?
3. Are there any real-world users beyond the author?
4. How does the tool handle data migration or integration with existing systems?
5. Is there a plan to move beyond local use, or to monetize this product?
6. What are the actual limitations of the OCR and identity matching logic in practice?