AI Stack Registry

Data notes

Artifacts are deterministic, versioned, and derived from listed package registries, provider docs, pip constraints semantics, and uv resolver docs. Current provider set: anthropic, google, openai, and xai. Each snapshot is dated and reproducible.

LLM training cutoffs mean model catalogs, version details, and defaults go stale quickly. Frontier providers update these frequently, so agents need a live source at runtime. Fetch /index.json, follow the lookup URLs, and browse /latest/models/index.md when a human needs the live catalog. The repo is private for now; the current public contract version is 0.2.0.

FAQ

Why does this exist?

LLMs have training cutoffs and can sound confident while being out of date. This registry keeps a continuously refreshed, public, machine-readable source of truth that stays current.

What does AI Stack Registry actually provide?

Dated snapshots plus a /latest/ alias of multi-provider model catalogs and dependency compatibility artifacts, so agents and humans can fetch current facts instead of guessing.

How do I use it with an agent?

Point the agent at https://aistackregistry.com/index.json, have it follow lookup.models_index, lookup.providers_index, and lookup.ecosystems_index, and treat latest_snapshot plus as_of as part of correctness. Use /latest/models/index.md when a human needs the readable catalog.

How does this reduce hallucinations?

Make the agent fetch authoritative metadata at runtime, require it to cite fetched artifact URLs in its output, and fail fast if the data cannot be retrieved rather than silently inventing.

What's intentionally out of scope (for now)?

This is not a framework tutorial site; it is a factual registry. Framework-level drift coverage may come later.