# Roadmap ## Current state V1 is complete and frozen. V2 is complete and archived. V2.1 is complete and archived. V3 is complete. V4 is complete. V5 is complete. V6 is complete. V7 is complete and archived. V7.1 is complete and archived. V7.2 is complete and archived. V8 is complete and archived. V9 is complete and archived. V9.1 is complete and archived. V10 is complete and archived. V11 is planned. V11.1 is active. V11.2 is planned. V11.3 is planned. V11.4 is planned. V11.5 is planned. V11.6 is planned. V11.7 is planned. V11.8 is planned. V11.9 is active. V11.10 is planned. V11.11 is planned. V11.12 is planned. ## V1 archive - [V1 phases](docs/planning/v1/phases.md) - [V1 next plan](docs/planning/v1/next-plan.md) - [V1 release gate](docs/planning/v1/release-gate.md) - [V2 plan](docs/planning/v2/plan.md) ## Current focus - keep the V1 release gate green - keep public docs aligned with shipped behavior - keep `ota agents` visible as a first-class derived adoption surface - keep the premium UX review loop green when text/help surfaces change - hold enterprise-facing scope behind the adoption readiness gate - keep the V11.1 execution-governance visibility and proof slice narrow and reviewable - keep the canonical roadmap aligned with the spec repo versioning ## V2 archive - [V2 plan](docs/planning/v2/plan.md) - [V2.1 plan](docs/planning/v2.1/plan.md) ## V6 archive - [V6 plan](docs/planning/v6/plan.md) ## V7 archive - [V7 plan](docs/planning/v7/plan.md) - [V7.1 plan](docs/planning/v7.1/plan.md) - [V7.2 plan](docs/planning/v7.2/plan.md) - [V8 plan](docs/planning/v8/plan.md) - [V9 plan](docs/planning/v9/plan.md) ## Active version - [V11.1 plan](docs/planning/v11.1/plan.md) ## Next planned version - [V11 plan](docs/planning/v11/plan.md) - [V11.2 plan](docs/planning/v11.2/plan.md) - [V11.3 plan](docs/planning/v11.3/plan.md) - [V11.4 plan](docs/planning/v11.4/plan.md) - [V11.5 plan](docs/planning/v11.5/plan.md) - [V11.6 plan](docs/planning/v11.6/plan.md) - [V11.7 plan](docs/planning/v11.7/plan.md) - [V11.8 plan](docs/planning/v11.8/plan.md) - [V11.9 plan](docs/planning/v11.9/plan.md) - [V11.10 plan](docs/planning/v11.10/plan.md) - [V11.11 plan](docs/planning/v11.11/plan.md) - [V11.12 plan](docs/planning/v11.12/plan.md) ## Implemented foundation - repo contract validation and execution - repo diagnosis and onboarding - conservative detection and init - workspace validation, diagnosis, bootstrap, and git-based acquisition ## Near-term next steps - preserve the shipped V10 semantic snapshot and correlation foundation - preserve the shipped repo/workspace trust baseline - keep docs and active planning aligned with the canonical spec repo - keep extension and editor surfaces contract-bound - avoid widening into generic plugin-runtime scope - stage V11 completion truth, reviewer evidence, and local/CI/agent convergence on top of the shipped semantic snapshot foundation - make V11.1 execution governance visibility and proof the active `1.6.23` slice - make V11.2 source convergence and detection governance the next disciplined widening slice after V11.1 - use V11.2 as the repo-truth convergence foundation before later execution-surface widening such as container-backed hydration and deterministic bootstrap materialization - make V11.3 agent-scoped execution enforcement the next runner slice after V11.2 so safe-task and workflow-closure truth become real runtime boundaries instead of review-only governance - make V11.4 the machine-readable governance output slice after V11.3 so CI, bots, and future control points can consume one canonical governance verdict instead of scraping runner output - make V11.5 the CI and merge-gate projection slice after V11.4 so contract-owned completion truth becomes enforceable at mandatory merge chokepoints - make V11.6 the harness and sandbox capability integration slice after V11.5 so external runners can enforce Ota’s callable boundary without re-deriving repo semantics - make V11.7 the audited execution boundary-crossing slice after V11.6 so allowed-but-heavier execution becomes explicit OSS evidence before enterprise approvals and waivers build on top - make V11.8 the sandbox policy compilation slice after V11.7 so contract-owned writable boundaries and egress posture can compile into real runtime enforcement targets instead of stopping at harness export - make V11.9 the governance truth reconciliation slice after V11.8 so emitted machine-readable governance stays faithful to the decision path that produced it - make V11.10 the replay-verified baseline trust slice after V11.9 so "last known good" becomes a stronger claim than one historical green witness - make V11.11 the machine-readable proof-boundary slice after V11.10 so honest narrow proof can publish what it covered and what it explicitly did not prove - make V11.12 the typed hydration input provenance slice after V11.11 so structured dependency hydration can name source/feed posture when that boundary materially changes trust ## Archived V2 shape Shipped V2 work so far is still intentionally narrow: - better real-repo detect coverage for common Node and Python repo shapes - conservative existing-contract comparison and additive merge for `ota detect` - stronger real-fixture coverage for mixed, legacy, and conflicting repo shapes - agent guidance surfaced on existing machine-readable and human-readable command paths ## Archived V2.1 shape The repo-first bridge work beyond that foundation focused on: - narrow interoperability framing without broad exports - stronger machine-facing guidance and JSON clarity - tighter workspace/team leverage boundaries without a second bootstrap engine ## Product direction V7 themes build on the shipped repo/workspace foundation with: - editor/IDE integration contract stabilization - remote runner metadata standard finalization - hosted validation workflow shape - stronger machine-facing operational summaries