This is the steering roadmap for Context Graph Protocol: from contextgraph/1.0-draft to a frozen contextgraph/1.0 with a real ecosystem around it. It came out of a full read of the three crates, the docs, the schema, and GOVERNANCE.md. Each item below is its own issue with the analysis, options, and acceptance criteria; this issue is the map.
The organizing idea
The protocol's identity is "enforced by contract, not convention". The roadmap is therefore ordered around closing every gap where the repo currently claims more than it enforces — because each one is a place where the project's own critique of the blob-pipe applies to itself. Three tracks:
Track 1 — Foundations (do first)
Track 2 — Protocol completeness (v0.2 milestone, all normative)
Every declared capability gets specified semantics or is dropped, and every documented contract gets an enforcement path — before the freeze, because the family-compatibility rule makes additions cheap after 1.0 but removals impossible.
Dead surface — spec or drop:
Underspecified semantics:
Claimed-but-unenforced:
Track 3 — The freeze and the ecosystem
contextgraph/1.0 freeze milestone:
Ecosystem milestone:
How this maps to the GOVERNANCE.md freeze criteria
| Freeze criterion |
Unlocked by |
| ≥2 independent implementations pass conformance |
#17 lowers the barrier; #18, #19 create the occasions; #3 makes it possible without reading Rust |
| 60-day stabilization window |
Finishing Track 2 starts the clock — every open normative issue resets it |
| No blocking normative issues |
Track 2 is exactly that list |
| Conformance fully enforces the documented requirements |
#8, #10, #12 (provider side); #14 (host side) |
Suggested sequencing
#2 → #16 immediately (foundations). Then the two decisions with the largest blast radius: #4 (wire framing) and #8 (token semantics), since most other normative items depend on or interact with them. The spec-or-drop trio (#5, #6, #11) can be decided quickly per the recommendation in each. #7 (graph) is the roadmap's centerpiece design work. Freeze-track items (#3, #14, #15) can proceed in parallel and are the last gate.
This is the steering roadmap for Context Graph Protocol: from
contextgraph/1.0-draftto a frozencontextgraph/1.0with a real ecosystem around it. It came out of a full read of the three crates, the docs, the schema, andGOVERNANCE.md. Each item below is its own issue with the analysis, options, and acceptance criteria; this issue is the map.The organizing idea
The protocol's identity is "enforced by contract, not convention". The roadmap is therefore ordered around closing every gap where the repo currently claims more than it enforces — because each one is a place where the project's own critique of the blob-pipe applies to itself. Three tracks:
Track 1 — Foundations (do first)
--misbehavered, schema validationTrack 2 — Protocol completeness (
v0.2milestone, all normative)Every declared capability gets specified semantics or is dropped, and every documented contract gets an enforcement path — before the freeze, because the family-compatibility rule makes additions cheap after 1.0 but removals impossible.
Dead surface — spec or drop:
upsert(capability negotiated, no wire method)subscribe(capability negotiated, no wire method)filters+ embedding-space negotiationUnderspecified semantics:
token_cost: canonical counting rule (today the budget check verifies arithmetic, not truth)Claimed-but-unenforced:
max_framesaudit, RFC 3339 temporal validation,as_ofprobeTrack 3 — The freeze and the ecosystem
contextgraph/1.0 freezemilestone:Ecosystemmilestone:How this maps to the GOVERNANCE.md freeze criteria
Suggested sequencing
#2 → #16immediately (foundations). Then the two decisions with the largest blast radius:#4(wire framing) and#8(token semantics), since most other normative items depend on or interact with them. The spec-or-drop trio (#5, #6, #11) can be decided quickly per the recommendation in each.#7(graph) is the roadmap's centerpiece design work. Freeze-track items (#3, #14, #15) can proceed in parallel and are the last gate.