Replace this file with the real plan for the crate generated from this template, or delete it if the project does not need a public roadmap.
Keep it short and honest: what exists, what is next, and what is deliberately out of scope. A roadmap that lists everything is a roadmap nobody trusts.
- module layout, crate-wide error type, and the public re-export surface
- lint configuration in
[lints], enforced identically locally and in CI - CI: format, clippy, build, test, per-file coverage, rustdoc, MSRV, and supply-chain checks
- a manual release workflow that versions, tags, publishes to crates.io, and creates a GitHub release with crate and TinyBus runtime/module assets
- a structured
mcp_calloutcome on every bridge call that reaches a result, for host metering and failure surfacing (contract 1.3) mcp.jsonreading for hosts with their own store (parse_with), and a guarded OAuth refresh on the flow- official-registry listings with one row per server at its latest version, the server's declared icon, per-request time budgets, and cached answers with a reported freshness when the registry stalls (contract 1.4)
- OAuth discovery for a 401 without
resource_metadata, from the origin's well-known protected-resource and authorization-server metadata - a local, background-synced index of the official catalog that answers searches instantly, and curated first-party servers with their endpoint, transport and authentication
- the first real feature area, replacing the placeholder
greetingmodule - module-level
README.mdanddocs/spec/entries as modules grow
- anything that cannot be tested deterministically
- convenience wrappers that hide the crate's error taxonomy from callers