Skip to content

ADR Index

ruv edited this page May 25, 2026 · 1 revision

Architecture Decision Records

This index lists all accepted ADRs (Architecture Decision Records) in the Ruflo project.


Critical Decisions (ADR-120+)

ADR-130: Graph Intelligence Integration

Status: Accepted | Impact: High Temporal knowledge graph with causal edges, PQ-compressed embeddings, and 6-algorithm pathfinding. Agents reason over decision chains with confidence decay and witness signatures (ADR-103 integration).

ADR-129: WASM Agent Composition + Managed Agents

Status: Accepted | Impact: High Dual execution model: rvagent (local WASM sandbox, <1ms overhead) + Anthropic Managed Agents (cloud). Agents compose from skills, support streaming.

ADR-127: GitHub Stack Modernization

Status: Accepted | Impact: Medium Unified GitHub integration (Actions, PRs, Issues) under single @claude-flow/github package. Removes duplication, adds swarm coordination.

ADR-126: Dual-Mode Orchestration (Claude Code + Codex)

Status: Accepted | Impact: Medium Claude agents and OpenAI Codex workers collaborate on the same task via shared memory. Each platform learns from the other.

ADR-125: ReasoningBank + AgentDB Unification

Status: Accepted | Impact: High All learning (patterns, trajectories, embeddings, feedback) lives in AgentDB with unified vector indexing. 150x-12,500x faster than JSON files.

ADR-124: Witness Temporal History (Extended)

Status: Accepted | Impact: Medium ADR-103 extended: witness signatures now include cryptographic proof of timestamp and causality. Enables compliance audits.

ADR-123: Sublinear Integration (Graph Intelligence v1)

Status: Accepted | Impact: High PageRank-weighted pathfinding on causal graphs. O(log n) instead of O(n²) for influence scoring. Enables 100-node swarms.


Federation & Distributed (ADR-107 to ADR-122)

ADR-120: Midstream QUIC From Agentic-Flow

Status: Accepted | Impact: High Replace hand-rolled QUIC with agentic-flow's native Rust binding. 40% latency reduction, automatic congestion control.

ADR-119: Midstreamer Adoption Assessment

Status: Accepted | Impact: Low Measure impact of Midstream adoption on federation speed. Result: 8% improvement on lossy networks, no regression.

ADR-117: Neural Trader — Managed Agent Backtests

Status: Accepted | Impact: Medium Long-horizon backtesting for autonomous trading agents. Uses Anthropic Managed Agents for live validation.

ADR-114: DSPy-TS Plugin

Status: Accepted | Impact: Low TypeScript port of Stanford DSPy for optimizing LLM calls. Integrated as optional plugin.

ADR-110: Production Spend Reporter

Status: Accepted | Impact: Medium Cost tracking across federation peers. Per-peer spend ledger with threshold alerts. ADR-097 Phase 3 completes here.

ADR-109: Receive-Side Dispatch

Status: Accepted | Impact: Medium Receiving agent decides how to route incoming messages (not sender). Enables adaptive load balancing.

ADR-108: Native QUIC Binding

Status: Accepted | Impact: High Rust QUIC binding for federation comms. 30% faster than Node.js TLS + HTTP/2. Replaces Node quic library.

ADR-107: Federation TLS

Status: Accepted | Impact: High Mutual TLS for agent-to-agent comms across machines. Certificates auto-generated per agent.

ADR-106: Peer Discovery

Status: Accepted | Impact: Medium mDNS + gossip for agent discovery on LAN. Falls back to explicit peer list for WAN.


Security & Core Foundations (ADR-90 to ADR-106)

ADR-103: Witness Temporal History

Status: Accepted | Impact: Critical Cryptographic manifest of all artifacts (code, models, decisions). Ed25519 signatures + SHA-256 hashes. Enables ruflo verify.

ADR-102: Input Validation @ Boundaries

Status: Accepted | Impact: Critical Zod schemas for all external inputs. Prevents injection attacks, type confusion. Tested via fuzzing.

ADR-101: Claims-Based Authorization

Status: Accepted | Impact: High Agents claim capabilities (file access, network, spawn children). Lead must grant before execution.

ADR-100: Cryptographic Agent Identity

Status: Accepted | Impact: High Each agent has Ed25519 keypair. Signs all messages. Enables federation trust and audit trails.

ADR-099: Dossier Investigator (Recursive Parallel Research)

Status: Accepted | Impact: Medium Multi-source entity investigation (web search, memory, codebase, git, ADRs). Parallel fan-out + recursive expansion.

ADR-098: Plugin Capability Sync

Status: Accepted | Impact: Medium Plugins declare required capabilities. CLI validates before load. Prevents unauthorized feature addition.

ADR-097: Federation Budget Circuit Breaker

Status: Accepted | Impact: High Distributed budget enforcement. Per-peer spending limits with threshold-based agent suspension. 3 phases (cost-track, cost-federation, spend-reporter).

ADR-096: Encryption at Rest

Status: Accepted | Impact: High All memory files encrypted via AES-256-GCM. Decryption happens in-process only.


Intelligence & Learning (ADR-085 to ADR-095)

ADR-094: Xenova → HuggingFace Transformers Migration

Status: Accepted | Impact: Medium Migrate from Xenova.js (ONNX in browser) to @xenova/transformers + local inference. Smaller bundle, same embeddings.

ADR-093: Mixture of Experts Routing

Status: Accepted | Impact: High Learned experts (agents) per domain. Thompson sampling-based selection. Cost-adjusted bandit.

ADR-092: SONA (Self-Optimizing Neural Architecture)

Status: Accepted | Impact: Critical Sub-millisecond adaptation via LoRA weights. Learns task complexity → agent type mapping. <0.05ms per routing call.

ADR-091: EWC++ (Elastic Weight Consolidation Plus)

Status: Accepted | Impact: High Prevent catastrophic forgetting when learning new patterns. Fisher information weighting.

ADR-090: ReasoningBank (v1)

Status: Accepted | Impact: Critical Centralized pattern storage with HNSW indexing. 150x faster search. Replaces 6+ per-agent caches.

ADR-088: LongMemEval Benchmark

Status: Accepted | Impact: Medium Benchmark long-context retrieval (100k token corpus). Ruflo scores 94% vs. LangChain 62%.

ADR-086: RuVLLM — Native Intelligence Backend

Status: Accepted | Impact: High Rust-backed embeddings + search. GPU optional. 75x faster with agentic-flow ONNX.


Early Decisions (ADR-1 to ADR-85)

Foundation and architectural decisions for Ruflo v3:

  • ADR-001 through ADR-085: Core domain boundaries, memory architecture, MCP surface, swarm coordination, hooks system, neural substrate, etc.

For full details, see v3/docs/adr/ directory.


Status Legend

Status Meaning
Accepted Approved and implemented
Proposed Under discussion
Superseded Replaced by newer ADR
Deprecated No longer used

Key Themes

Multi-Agent Orchestration

ADR-001, 003, 005, 007, 025, 026, 028, 090, 092, 093, 123

Security & Trust

ADR-096, 100, 102, 103, 104, 124, 127

Federation & Distribution

ADR-097, 106, 107, 108, 109, 110, 120, 127

Learning & Intelligence

ADR-086, 088, 090, 091, 092, 093, 113, 125

Graph & Knowledge

ADR-123, 125, 130


Ruflo v3.10.1 · 130+ ADRs · GitHub

Clone this wiki locally