Migrated from Method backlog

This issue was created from a legacy filesystem backlog card. GitHub Issues are now the live work tracker; repository docs remain Method evidence.

Source backlog: docs/method/backlog/asap/PLATFORM_echo-contract-hosting-roadmap.md
Original lane: asap
Original legend: PLATFORM

Original backlog card

Echo Contract Hosting Roadmap

Status: active sequencing card.

Echo should become a generic host for Wesley-compiled contract families. It
must not grow application-specific APIs for text editing, code intelligence,
debugging, or any other consumer domain.

Doctrine

Echo is the deterministic witnessed causal substrate.

Wesley authors and compiles contract families from GraphQL into generated Rust,
ABI codecs, schema identity, operation catalogs, and app-level helpers.

Applications such as jedit own their domain contracts and product behavior.
Echo already exposes the generic WASM substrate: EINT dispatch_intent(...),
observe(...), registry metadata exports, and app-agnostic kernel boundaries.
The roadmap must extend that existing path, not invent a second envelope or
registry model.

Sequence

1. Wesley compiled contract hosting doctrine
   • Design packet:
     0013 - Wesley Compiled Contract Hosting Doctrine
2. Existing EINT, registry, and observation boundary inventory
   • Completed design packet:
     0014 - EINT, Registry, And Observation Boundary Inventory
3. Registry provider wiring and host boundary decision
   • Completed design packet:
     0015 - Registry Provider Host Boundary Decision
4. Wesley to Echo toy contract proof
   • Status: accepted.
   • Design packet:
     0016 - Wesley To Echo Toy Contract Proof
   • Retro:
     0016 - Wesley To Echo Toy Contract Proof
5. Installed package registry, mutation dispatch, query observer bridge, and
   local installed intent pipeline
   • Status: complete in core/local fixtures.
   • Backlog checkpoints:
     Installed Wesley contract host dispatch,
     Contract QueryView observer bridge
6. Contract-aware receipts and readings
7. Contract reading identity and bounded payloads
8. Contract artifact retention in echo-cas
9. Contract retention and semantic lookup seams
10. External contract proof fixture
11. jedit text contract MVP
12. Graft live frontier structural readings
13. Contract strands and counterfactuals
14. Continuum contract artifact interchange
15. Contract-hosted file history substrate
    • Design packet:
      0018 - Contract-Hosted File History Substrate

Non-goals

• Do not add ReplaceRange, BufferWorldline, or text-editing types to Echo
  core unless they are generated application contract payloads.
• Do not add a special jedit ABI.
• Do not create a second intent envelope when EINT already exists.
• Do not create a second registry model when echo-registry-api and
  echo-wesley-gen already exist.
• Do not let Graft mutate Echo state directly.
• Do not build dynamic plugin loading before static contract hosting works.
• Do not start IPA, proof systems, or network Continuum protocol work in this
  cluster.

Done looks like

• Each item in the sequence has a narrow card with dependencies, acceptance
  criteria, and non-goals.
• Future agents can pick the next card without re-arguing whether Echo owns
  application-specific APIs.