Fast US Employer Health Plan Pricing for AI Agents

The description states that HealthPorta is a system designed to convert large, nested JSON files containing publicly disclosed U.S. employer health-plan negotiated rates into structured, queryable formats for AI agents and applications.

It includes:

• A Rust-based scanner and finalizer for processing compressed payer files.
• Python orchestration for discovery, publication, auditing, and activation.
• Immutable, content-addressed PostgreSQL serving blocks.
• Dense identifiers, shared price dictionaries, and query-oriented indexes.
• Support for both MCP and REST APIs.
• A separate logical employer-plan binding system, allowing reuse of identical physical data while preserving plan ownership.
• An audit-before-activation release gate.

The system is described as being capable of importing a 292GB JSON file in 8m54s, reducing it to a 2.48GB query-ready payload.

The description states: “HealthPorta turns publicly disclosed employer-plan data into a Find Care workflow that MCP-compatible agents and applications can use through MCP or REST APIs.”

The project positions itself as an infrastructure tool for making U.S. Transparency in Coverage data usable by AI agents, particularly those built with OpenAI’s tools like Codex and GPT-5.6.

It claims to solve a problem where:

• Publicly available employer-plan pricing files are too large and complex for AI systems to process.
• The system enables workflows such as identifying relevant employer plans, resolving care needs into procedures, finding nearby providers, retrieving negotiated rates, and comparing options with source provenance.

The author notes that this is not about providing personalized estimates or guarantees of coverage but rather about making disclosed negotiated rates accessible for agent-based decision-making.

The description states: “We wanted an agent to handle a practical request such as: My employer’s plan covers me, and I need care near me. Which in-network providers have disclosed negotiated rates for this service?”

It also mentions that the system was built during OpenAI Build Week, leveraging AI tools like Codex and GPT-5.6.

The description states: “Working with Codex and GPT-5.6, we built and hardened...”

The target customer appears to be AI developers or teams building agent-based applications that need access to structured employer health-plan pricing data.

The system is positioned for:

• MCP-compatible agents
• Applications using REST APIs
• Systems needing fast, auditable access to negotiated rates

There is no indication of specific end-users (e.g., consumers, healthcare providers, insurers) or direct integrations beyond the described API interfaces.

The description states: “HealthPorta turns publicly disclosed employer-plan data into a Find Care workflow that MCP-compatible agents and applications can use through MCP or REST APIs.”

No business model or pricing information is provided in the description.

The system is presented as an open-source component (healthcare-mrf-api) and part of a larger pipeline, but there is no mention of monetization strategies, licensing terms, or commercial offerings.

The description states: “HealthPorta (open source part: healthcare-mrf-api) and our API and MCP delivery layers existed before OpenAI Build Week.”

Key technical signals from the description include:

• Use of Rust for scanning large files.
• Python orchestration for discovery, publication, auditing, and activation.
• PostgreSQL with immutable, content-addressed serving blocks.
• Support for MCP and REST APIs.
• Implementation of audit-before-activation gates.
• Separation of physical file identity from logical plan identity, enabling reuse of data without mixing ownership.
• Use of Codex and GPT-5.6 in the engineering loop to trace bottlenecks, evaluate alternatives, implement improvements, and verify deployment.

The description states: “We used Codex to trace bottlenecks across a large multi-stage pipeline.”

There is no evidence of traction or maturity beyond the author’s own development work during OpenAI Build Week.

The system is described as:

• Having been developed during a hackathon.
• Being in an early stage of benchmarking and unit-economics measurement.
• Not yet integrated into production systems or widely adopted.

The description states: “This is one measured case, not a claim that every payer file will produce the same result.”

No competitive landscape or direct competitors are mentioned in the description.

The project does not reference other tools or platforms dealing with health plan pricing data or transparency rules.

The description states: “What existed before Build Week...” — implying prior existence but no mention of existing solutions or market players.

Key risks and red flags include:

• No evidence of real-world usage or adoption.
• Lack of revenue, customer, or traction data.
• Unverified claims about performance metrics (e.g., 8m54s import time).
• No indication of scalability beyond one measured case.
• Unclear long-term viability or commercialization strategy.
• Reliance on AI tools like Codex and GPT-5.6, which may not be available in all contexts.

The description states: “Everything above is the authors' own account. It is not independently verified, and no revenue, customer or traction data is available beyond what they state.”

1. Has HealthPorta been used by any third-party developers or organizations outside of the Build Week project?
2. Are there any existing integrations with healthcare systems, AI agents, or applications using the API?
3. What is the current status of unit economics benchmarking and storage cost modeling?
4. How does the system handle updates to monthly Transparency in Coverage files?
5. Is there a plan for monetization or commercial deployment beyond the open-source components?
6. Can you provide more details on how the audit-before-activation process works in practice?