Regstair: OCI Routing Fabric

The description states that Regstair is an OCI routing fabric designed to provide a single logical endpoint for container image operations (pull/push) across internal and external registries. It functions as an abstraction layer, separating the logical identity of images from their physical storage locations.

It includes components such as:

• An OCI Gateway handling client requests
• A Route Engine matching repository names against policy rules
• Registry Connectors isolating upstream-specific behavior
• A Content-Addressed Cache storing objects by digest
• Metadata and Provenance tracking

The system supports:

• Deterministic pull routing
• Push routing to defined destinations
• Namespace authority enforcement
• Fallback control (per namespace)
• Digest caching with reuse of shared layers
• Provenance logging for each operation
• Human-readable explanations of routing decisions

Not evidenced: actual deployment, usage metrics, or integration with real registries beyond the author’s test environment.

The description frames Regstair as:

• A registry abstraction layer, not a replacement for existing registries.
• An OCI-compliant interface that allows clients to interact without knowing where images live.
• A tool for separating logical identity from physical registry location.
• A solution to brittle dependencies caused by registry leakage into workflows.

It positions itself as more than a pull-through cache, emphasizing:

• Policy-based routing
• Namespace authority
• Fallback control
• Logical naming
• Provenance tracking

Claims are self-reported and unverified. No external validation or market positioning data is provided.

The description implies Regstair targets organizations using multiple container registries, particularly those suffering from:

• Registry leakage in workflows (Dockerfiles, CI pipelines, Kubernetes manifests)
• Brittle dependencies due to registry changes
• Outages or rate limits affecting builds
• Dependency-confusion risks from unsafe fallback behavior

It appears aimed at enterprises managing complex multi-registry environments where:

• Internal and external registries coexist
• Developers need stable image names while storage locations change
• Security and governance require explicit namespace ownership

Not evidenced: specific customer segments, personas, or use cases beyond the author’s own scenario.

There is no evidence of a business model or pricing structure. The description only mentions:

• The project was built during a hackathon (Devpost submission)
• It is described as a vertical slice for demonstration purposes
• No revenue, funding rounds, or monetization strategy are stated

Not evidenced: any commercial offering, pricing tiers, or customer acquisition plans.

The description indicates:

• Built using Go, Docker, SQL, REST APIs, HTTP
• Deliberately separated components (Gateway, Route Engine, Connectors, Cache)
• Supports OCI Distribution protocol semantics
• Implements digest-based caching and reference-aware cleanup
• Includes an administrative web interface
• Uses AI tools (Codex, GPT-5.6) for development assistance but not live enforcement

It also notes:

• Challenges in handling mutable tags, multi-platform images, authentication
• Limitations in early version: no replication, signature verification, SBOMs, etc.
• Focus on proving core abstraction rather than full platform

Not evidenced: production-grade reliability, scalability, or performance benchmarks.

The description states:

• This is a Build Week hackathon project
• No revenue, customers, or traction data are available
• It was submitted to the OpenAI 2026 hackathon on Devpost
• The author describes it as a “vertical slice” demonstrating core functionality

No evidence of:

• User adoption
• Customer feedback
• Product-market fit
• Iteration history or roadmap beyond initial demo

The description does not mention competitors directly. However, it implies Regstair operates in the space of:

• Container registry abstraction layers
• Pull-through caches (e.g., Harbor, Quay, Zot)
• OCI-compliant registries and proxy solutions

It distinguishes itself from existing pull-through caches by focusing on:

• Policy-based routing
• Namespace authority
• Provenance tracking
• Deterministic decision-making

Not evidenced: competitive landscape analysis, market share, or differentiation from similar tools.

Key risks and red flags based on the description:

• No real-world usage: The project is presented as a hackathon demo with no evidence of adoption.
• Unproven maturity: No production deployments, performance data, or long-term stability.
• Limited scope: Early version focused only on core abstraction; many advanced features (e.g., replication, vulnerability scanning) were not implemented.
• AI dependency: While AI is used for policy generation and explanation, enforcement remains deterministic — but this raises questions about how much of the system relies on human oversight.
• Self-reported nature: All claims are unverified and self-descriptive; no third-party validation or external feedback.

1. What is the actual problem you're solving in practice? Is there a real-world organization using this?
2. How does Regstair handle large-scale registry traffic or high availability requirements?
3. Can you walk us through how a typical enterprise would integrate Regstair into their existing workflow?
4. What are the key assumptions behind your routing policies, and how do they scale?
5. Are there any known edge cases or failure modes in the current implementation?
6. How do you plan to evolve beyond the current vertical slice?
7. Do you have any plans for monetization or commercialization?