Zion Scout for Camping Leads

Project: Zion Scout for Camping Leads

Author's Claim: A tool that turns natural-language sales strategies into ranked opportunity maps and human-reviewed outreach using AI.

What Changed: The project description shows a self-contained, hackathon-built product with an AI agent that interprets sales intent in natural language and translates it into actionable CRM signals. It is not evidenced to have traction, revenue or customers.

Single Most Important Open Question: Is the described AI workflow actually grounded in real data and capable of being scaled beyond a demo?

The description states that Zion Scout is a Next.js application backed by Supabase, PostgreSQL, and PostGIS, with a server-only route that sends sanitized CRM signals to the GPT-5.6 Responses API. It uses strict Structured Outputs to enforce recommendation schema before model output becomes interface state.

It is described as an AI agent that analyzes a bounded CRM snapshot and returns a ranked shortlist with evidence, which then narrows a geospatial map to those opportunities, shows why each property was selected, and drafts personalized outreach for review. The user can inspect the underlying lead, copy the draft, and update CRM status without leaving the workflow.

The system is designed to not expose contact details or API credentials in the browser-to-model payload, and it uses read-only recommendations with human approval before outreach.

• Evidenced: Yes
• Inferred: No

The author states that Zion Scout addresses a common problem for small hospitality sales teams: they have hundreds of leads but still struggle to answer three basic questions: Who should I call today? Why are they a fit? What should I say?

It is positioned as a tool that compresses a manual workflow into one explainable, human-approved interaction. The product is described as not just a chat wrapper, but as an AI that changes product state, turning natural-language intent into a ranked map, evidence cards, and a concrete next action.

The author also claims that the system is built to ground every recommendation in supplied sales signals and that it keeps API credentials and contact details out of the browser-to-model payload.

• Evidenced: Yes
• Inferred: No

The description states that Zion Scout targets small hospitality sales teams, particularly those with hundreds of leads who struggle to prioritize outreach. It is designed for salespeople who need to make decisions about which leads to contact, why they are a fit, and what to say.

It is not evidenced whether the product is intended for a specific vertical beyond camping or glamping, nor whether it targets a specific size of team or company.

• Evidenced: Yes
• Inferred: No

There is no evidence in the description of a business model or pricing structure. The project is described as a hackathon submission and not as a commercial product with revenue streams, subscriptions, or pricing tiers.

• Evidenced: No
• Inferred: No

The system is built using:

• Next.js
• Supabase, PostgreSQL, PostGIS
• GPT-5.6 Responses API
• Codex for development acceleration
• Leaflet.js, OpenStreetMap, PostgreSQL

It uses strict JSON Schema output, server-side ID validation, and bounded inputs to ensure that model outputs are safe, grounded, and actionable.

The system is designed to not expose contact details or API credentials in the browser-to-model payload, and it includes RLS policies and automated security tests for the underlying CRM platform.

• Evidenced: Yes
• Inferred: No

There is no evidence of traction, revenue, customers, or adoption. The project is described as a hackathon submission, and no data on usage, retention, or product-market fit is provided.

• Evidenced: No
• Inferred: No

The description does not mention any competitors or direct market context. It is unclear whether similar tools exist in the marketplace for sales lead prioritization or AI-powered CRM workflows.

• Evidenced: No
• Inferred: No

1. The system is built as a hackathon submission, not a commercial product, and there is no evidence of traction or scalability.
2. The use of GPT-5.6 implies a proprietary or high-end model, but the description does not clarify access or cost.
3. The system relies on sanitized CRM data, which may not reflect real-world data quality or integration challenges.
4. There is no evidence of human-in-the-loop feedback loops or measurable impact on sales outcomes.
5. The project is self-reported and unverified, with no third-party validation.

• Evidenced: Yes
• Inferred: No

1. What is the actual CRM data source, and how does it integrate with the system?
2. How is the AI model trained or fine-tuned for this specific use case?
3. Is there any evidence of human-in-the-loop feedback or impact measurement?
4. What are the real-world constraints or limitations of using this in a production sales environment?
5. Are there plans to scale beyond a demo, and what would that look like?

• Evidenced: No
• Inferred: Yes