ruvnet / ruvnet/RuVector

Release: RuVix Cognition Kernel — ADR-087 & Implementation Plan

Open
#249 1 comment 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

enhancement
Dominant language
Rust
Stars
4.5k
Forks
603
Avg merge
23h 32m
Merged PRs (30d)
59

Description

Overview

ADR-087 defines RuVix — a cognition kernel for the Agentic Age built on RuVector and RVF primitives. This issue tracks the release of the architecture document and subsequent implementation.

PR: #248
ADR: ADR-087-ruvix-cognition-kernel.md

What RuVix Is

A minimal cognition kernel with 6 primitives (task, capability, region, queue, timer, proof) and 12 syscalls. Not a Linux clone — a purpose-built OS for agentic workloads where:

  • RVF is the boot object — signed packages are complete cognitive units
  • RuVector is kernel-resident — vectors and graphs are native kernel resources, not userspace databases
  • Every mutation is proof-gated — kernel invariant, not library convention
  • Scheduling is coherence-aware — deadline + novelty signal + structural risk
  • No POSIX — WIT/RVF contracts replace files, processes, and pipes

Syscall Surface (12 total)

task_spawn, cap_grant, region_map, queue_send, queue_recv, timer_wait,
rvf_mount, attest_emit, vector_get, vector_put_proved, graph_apply_proved,
sensor_subscribe

Build Path

Phase A: Linux-Hosted Nucleus (Days 1-60)
Crate Purpose Weeks
ruvix-types Kernel types (no_std, no alloc) 1-2
ruvix-region Slab/append-only/immutable regions 3-4
ruvix-queue io_uring-style ring buffer IPC 5-6
ruvix-cap Capability table + derivation tree 7-8
ruvix-proof Proof token generation/verification 9-10
ruvix-vecgraph Kernel vector/graph stores 11-12
ruvix-sched Coherence-aware scheduler 13-14
ruvix-boot RVF boot loader 15-16
ruvix-nucleus Integration + ABI freeze 17-18
Phase B: Bare Metal AArch64 (Days 60-120)
  • AArch64 bootstrap (exception vectors, MMU, UART)
  • Region → physical memory mapping
  • GIC interrupt → queue delivery
  • WASM component runtime on bare metal
  • Hardware scheduler with timer preemption

Demo Applications (8 total)

# Demo Category
1 Proof-Gated Vector Journal Foundational
2 Edge ML Inference Pipeline Practical
3 Autonomous Drone Swarm Coordinator Practical
4 Self-Healing Knowledge Graph Practical
5 Collective Intelligence Mesh Exotic
6 Quantum-Coherent Memory Replay Exotic
7 Biological Signal Processor Exotic
8 Adversarial Reasoning Arena Exotic

Acceptance Test

A signed RVF boots, consumes a RuView event, performs one proof-gated vector mutation, emits an attestation, and replays to the same state after restart.

Integration with Existing Crates

Maps 12 existing crates: ruvector-core, ruvector-graph-transformer, ruvector-verified, cognitum-gate-kernel, ruvector-coherence, ruvector-mincut, rvf, ruvector-raft, ruvector-snapshot, sona, ruvllm, ruvector-temporal-tensor.

Comparison

Property Linux seL4 RuVix
Kernel LOC ~30M ~8.7K <15K target
Syscalls ~450 ~12 12
Vector/graph native No No Yes
Proof-gated mutation No No Yes (kernel invariant)

Release Checklist

  • ADR-087 merged (#248)
  • ruvix-types crate scaffolded (no_std, no alloc)
  • Phase A crate workspace structure created
  • Demo 1 (Proof-Gated Vector Journal) spec finalized
  • CI pipeline for no_std builds
  • ABI documentation generated from ruvix-types

Contributor guide

No contributing guide indexed for this repository

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start by reading docs/adr/ADR-087-ruvix-cognition-kernel.md and the release checklist in this issue. Review the planned ruvix-* crate workspace and existing integration crates before choosing a scoped implementation task. Done requires the relevant Phase A scaffolding, no_std CI, generated ABI documentation, and the signed-RVF boot acceptance flow.

Written by the indexing model from the issue text.

Assessment

Tech stack
rust, wasm
Domain
ai, operating-systems
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
15/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.