ruvnet/ruflo
51.3
Adequate · 14 November 2026
703.8k
lines of production code
TypeScript
with JavaScript
6
measurements over time
What this system is
Features
AIDefence mod: secret and injection guard for learning samples
The ruflo-aidefence plugin now includes a 'mod' that acts as a tighten-only guard on what the \aidefence\_learn\ tool teaches the detector. This prevents secrets (keys, tokens, passwords) and prompt-injection phrasing from being stored in learned patterns. The mod registers a \/aidefence-mod\ command for status and scanning, and writes session statistics to a status file for the console to read.
plugins/ruflo-aidefence/hooks · high confidence
AgentDB memory backend with retrieval security and scoped storage
The memory package now includes a new AgentDB backend (agentdb-backend) that provides HNSW-based vector search with graceful fallback to WASM or in-memory storage, and a new AgentDB adapter (agentdb-adapter) that fixes same-key upserts to prevent orphaned entries and stale search results. A new AgentDbRetrievalGuard (ADR-377) filters retrieval results for injected or oversized content, controlled by the CLAUDE\_FLOW\_RETRIEVAL\_GUARD and CLAUDE\_FLOW\_RETRIEVAL\_GUARD\_STRICT environment variables. Additionally, a new agent-memory-scope module introduces project, local, and user-scoped memory directories with path sanitization and knowledge transfer capabilities.
v3/@claude-flow/memory · high confidence
Autopilot Mod: Guard against secrets and injection in agent writes
A new autopilot mod plugin has been added that acts as a tighten-only guard on the agent's log and learned patterns (\autopilot\_log\, \autopilot\_learn\) and enforces a runaway loop limit on \autopilot\_config\ (\maxIterations\). The mod intercepts tool calls to these specific functions and refuses writes that contain detected secrets (such as API keys, tokens, or passwords) or prompt-injection phrasing, ensuring only references are stored. It also blocks \maxIterations\ values that exceed a configurable cap. Users can manage the mod via the \/autopilot-mod\ command (status, scan, limits) and view session statistics in a status file.
plugins/ruflo-autopilot/hooks · high confidence
Browser mod adds security guard and status commands
The ruflo-browser plugin now includes a tighten-only security guard that intercepts browser\\ tool calls to prevent cookie or storage exfiltration, block hostile URLs (such as cloud metadata services or plain HTTP outside localhost), and reject unsafe Chrome launch flags. It also introduces a local /browser-mod command for checking guard status, scanning scripts for exfiltration risks, and validating URLs, along with a status file that tracks blocked calls for the console.
plugins/ruflo-browser/hooks · high confidence
Business pods mod with secret-screening guard and dry-run sales pod execution
The ruflo-business-pods plugin now ships as a mod (ADR-445 pattern) that includes a tighten-only tool guard to prevent secrets and credentials from entering pod templates or template paths, a local slash command (/pods-mod) for status, scanning, and path validation, and a status file for console visibility. It also provides a single-tick runner (pod-tick.mjs) for the sales pod template that validates templates, resolves agent roles, reserves budget via a file-based stub ledger, constructs dry-run prompts, and posts an envelope summary to the room, with live execution reserved for Phase 3.
plugins/ruflo-business-pods · high confidence
ChatGPT app submission reference materials added
The \chatgpt-plugin\ directory now contains the complete set of documentation and submission artifacts for the RuFlo Federation ChatGPT app, including the structured submission JSON, portal listing copy, reviewer test cases, an application runbook, and six app-specific skills. These files serve as the reviewable source of truth for preparing the OpenAI platform submission and do not alter the production MCP implementation, which remains in the \plugins/ruflo-x-gateway\ directory.
chatgpt-plugin · high confidence
Claude Flow Plugin installation and hook scripts
The plugin now ships with a set of shell and Node.js scripts in .claude-plugin/scripts to manage the user experience. The install.sh script handles prerequisites (Claude Code CLI, Node.js \>= 20), allows users to choose which components to install (commands, agents, or MCP servers), and configures the MCP server entries in settings.json. The uninstall.sh and verify.sh scripts provide corresponding cleanup and health-check capabilities. Crucially, the hook invokers (ruflo-hook.sh and ruflo-hook.cjs) have been hardened: they prefer locally installed binaries over repeated npx calls to avoid install failures and slow-downs, enforce a 15-second timeout to prevent hanging processes, and ensure hooks always exit 0 so they never block the user's turn. The .cjs shim specifically addresses Windows compatibility by bypassing shell execution for better reliability and security.
.claude-plugin/scripts · high confidence
Decentralized plugin store with IPFS discovery and search
The CLI now includes a new plugin store module that enables discovering, searching, and managing plugins via a decentralized IPFS-based registry. Users can browse official and community plugins, filter by category, tags, trust level, and permissions, and view trending or featured plugins. The store verifies registry signatures using Ed25519 to ensure integrity and supports caching for faster access.
v3/@claude-flow/cli/src/plugins/store · high confidence
Federation mod: secret and PII guard with status and scan commands
The ruflo-federation plugin now includes a local mod that enforces a security guard on outbound federation tool calls. When enabled (the default), the mod intercepts specific publish and sync tools to scan their inputs for secrets (such as API keys, tokens, and passwords) and personally identifiable information (SSNs and credit card numbers); if detected, the call is denied with a reason. A new \/federation-mod\ command allows users to check the guard's status, list connected federation tools, or scan arbitrary text for potential leaks. The mod also writes a session status file to track how many calls were blocked.
plugins/ruflo-federation/hooks · high confidence
Flash Attention integration and resource admission policy
The performance module now includes a Flash Attention integration that wraps the @ruvector/attention runtime, providing optimized attention classes (FlashAttention, DotProductAttention, MultiHeadAttention, LinearAttention) and a FlashAttentionOptimizer for benchmarking and metrics tracking. A comprehensive benchmarking framework has been added to validate performance targets (e.g., 2.49x-7.47x speedup) and track memory usage. Additionally, a task-conditioned resource admission policy has been implemented to decide whether to admit, defer, or reject tasks based on resource profiles, host snapshots, and configurable policies, with results exported for use by other system components.
v3/@claude-flow/performance · high confidence
Initial agent configuration and skill definitions for OpenAI Codex CLI
This change introduces the \.agents\ directory structure, providing the foundational configuration and skill definitions for the OpenAI Codex CLI. It adds a \config.toml\ file that sets the default model to \gpt-5.3-codex\, configures approval policies, sandbox modes, and MCP server connections, and defines specific profiles for development, safe, and CI environments. Additionally, it populates the \skills/\ directory with a comprehensive suite of agent skills—including adaptive coordination, goal planning, code quality analysis, system architecture, and payment authorization—each defined by \SKILL.md\ files that specify triggers, capabilities, and tool integrations.
.agents · high confidence
Initial release of the GitHub clone and npm download tracking ledger
The data directory now includes a structured ledger (clone-data.ledger.json) and a proof manifest (clone-data.proof.json) that track 14-day rolling GitHub clone counts and 12-month npm download totals for the ruvnet ecosystem repositories (ruflo, agentdb, agentic-flow, ruvector, ruv-FANN). This change introduces the underlying data schema and snapshot history required for the clone tracker feature, establishing the source of truth for the '11.7M npm downloads' badge and clone metrics displayed in the project.
data · high confidence
Introduce @claude-flow/cli-core as a lightweight, standalone memory and MCP tool-definition module
This change introduces the \@claude-flow/cli-core\ package, a lightweight core module designed to provide memory persistence and MCP tool definitions without the heavy dependencies of the full CLI. It includes a \JsonMemoryBackend\ that persists data to a local JSON file (defaulting to \.swarm/memory.json\), supporting key-value storage, retrieval, listing, deletion, and substring-based search, along with a \memory\ CLI subcommand for direct interaction. Additionally, it exposes pure data definitions for \memory\\\ and \hooks\\\ MCP tools, allowing external MCP servers to advertise these capabilities without importing the full handler logic, which remains in the main \@claude-flow/cli\ package. The package also includes input validation utilities for security-sensitive fields like environment variables and file paths.
v3/@claude-flow/cli-core · high confidence
Introduce @claude-flow/teammate-plugin for multi-agent orchestration
The new @claude-flow/teammate-plugin bridges Claude Code v2.1.19+ multi-agent capabilities with Claude Flow, enabling users to spawn and manage teams of AI teammates. It exposes 21 MCP tools for team management, messaging, and plan approval, and includes a semantic router and topology optimizer that leverage WebAssembly (via @ruvnet/bmssp) for intelligent task routing and graph-based communication optimization. The plugin also provides built-in resilience features like rate limiting, circuit breaking, health checking, and retry logic to ensure stable multi-agent execution.
v3/plugins/teammate-plugin · high confidence
Introduce AIDefence AI manipulation detection and self-learning service
The AIDefence module now provides a unified facade for detecting AI manipulation attempts (such as prompt injection, jailbreaks, and role-switching) using over 50 built-in regex patterns, alongside a self-learning service that tracks mitigation effectiveness and supports vector-based pattern search. Users can enable optional PII detection and configure a custom vector store (defaulting to an in-memory implementation) to integrate with external systems like AgentDB for faster similarity searches.
v3/@claude-flow/aidefence · high confidence
Introduce Cognitum IoT plugin with device lifecycle, telemetry, and fleet management
The @claude-flow/plugin-iot-cognitum package now provides a complete IoT device management layer. It introduces a CLI (cognitum-iot) for registering, pairing, and querying devices, with authentication via the COGNITUM\_SEED\_TOKEN environment variable. The plugin exposes an IoTCoordinator that orchestrates device lifecycle (registration, pairing, trust scoring), telemetry ingestion and vector-store querying, anomaly detection, and fleet topology management. It also includes a state-machine-based firmware orchestration service supporting canary, rolling, and blue-green deployment strategies with automatic rollback on anomaly thresholds.
v3/@claude-flow/plugin-iot-cognitum · high confidence
Introduce OAuth 2.1 authentication and secure Nostr publishing for the ChatGPT Federation plugin
The ChatGPT Federation plugin now implements an OAuth 2.1 resource server (RFC 9728) to secure its MCP tools, replacing the legacy \x-caller-token\ mechanism. Access tokens are validated against Cognitum's authorization server, with audience claims bound to the client ID to prevent cross-app token misuse. The plugin also introduces a dedicated, authenticated WebSocket connection to the Nostr relay for publishing events, ensuring that the connector signs and publishes messages under its own verified identity without exposing secret keys to the model or external callers.
plugins/ruflo-chatgpt-federation/src · high confidence
Introduce RuVector WASM integration bridges for advanced AI capabilities
Added a new plugin location providing integration bridges for 15 upstream RuVector WASM packages, enabling features such as ultra-fast vector similarity search (HNSW), flash attention, graph neural networks, hyperbolic embeddings, reinforcement learning, quantum-inspired optimization, and cognitive kernels. The implementation includes a centralized WASM registry for lifecycle management and provides mock fallbacks for development environments.
v3/plugins/ruvector-upstream · high confidence
Introduce Rust-based federation peer with in-process safety gating
The new \ruflo-federation-peer\ crate replaces the previous Node-bridge path with a single Rust process that handles federation hops and in-flight safety scanning. It integrates \midstreamer-quic\ for transport and \aimds-core\ for the 3-gate safety pipeline (detection, analysis, response), allowing operators to run a peer that inspects, blocks, or redacts messages before dispatching them to the local agent via stdio. The crate exposes pluggable traits (\TransportProvider\, \SafetyGate\, \Dispatcher\) so downstream consumers can substitute their own backends, while the default \native\ feature provides a ready-to-run binary for production use.
v3/crates/ruflo-federation-peer · high confidence
Introduce Witness toolkit for cryptographically signed fix verification and temporal history
The witness scripts in \plugins/ruflo-core/scripts/witness\ now provide a complete toolkit for verifying that code fixes are present in a repository. The new \init.mjs\ bootstraps a project with a manifest and history file, while \regen.mjs\ generates a signed manifest using Ed25519 signatures and \@noble/ed25519\. The \verify.mjs\ script checks the live tree against the manifest, supporting both source-only and full-tree verification, and correctly handles missing build artifacts as a precondition failure rather than a regression. Additionally, \history.mjs\ allows querying the temporal history of fixes to identify regressions, and \perf.mjs\ benchmarks key capabilities like install time and memory load to detect performance regressions.
plugins/ruflo-core/scripts/witness · high confidence
Introduce dedicated ChatGPT Federation service with isolated identity and OAuth 2.1 auth
The ChatGPT Federation connector is now a standalone Cloud Run service that holds its own Nostr signing key and opens its own authenticated WebSocket connection to the relay, ensuring the service's identity matches the authenticated NIP-42 identity required by the relay. Authentication for publishing is handled via OAuth 2.1 against Cognitum's authorization server using public-client + PKCE (no client secret), with access tokens bound to the client ID rather than the resource URL. The service enforces strict key custody by mounting the signing key from Secret Manager as a read-only file, pinning the secret version to prevent silent identity rotation, and refusing to start if the derived public key does not match the expected identity.
plugins/ruflo-chatgpt-federation · high confidence
Introduce goal-driven research UI with GOAP planning and AI-assisted workflows
The v3/goal\_ui area now provides a complete interface for defining, planning, and reviewing research objectives using a Goal-Oriented Action Planning (GOAP) system. Users can input research goals, select from domain-specific presets (e.g., Finance, Medical, Coding), and trigger AI-generated goal optimization via Supabase functions. The interface visualizes the planning process through animated step components (AgentStep, DevelopmentStep) that track execution states (pending, active, completed) and display real-time progress. A dedicated StateAssessmentCard allows users to monitor current vs. goal states, while the ResearchReviewCard facilitates final approval or revision requests before development. The system also includes a GOAPConfigDisplay for viewing execution modes and replanning triggers, and a WidgetCustomizer for branding and embedding the research tool.
_v3/goal\ui · high confidence
Introduce ruflo-daa secret guard and /daa-mod diagnostic command
The ruflo-daa plugin now includes a session-scoped secret guard that intercepts calls to tools prefixed with \daa\\ and blocks them if the input contains detected secrets (such as API keys, tokens, or passwords) or injection patterns, returning a refusal message instead. A new \/daa-mod\ command provides local diagnostics without model turns, allowing users to check the guard's on/off status, view blocked call counts, scan arbitrary text for potential secrets, and list connected \daa\\*\ tools. The guard's behavior is configurable via a \guard\ option (defaulting to on) and session statistics are persisted to a status file.
plugins/ruflo-daa/hooks · high confidence
Introduce ruflo-mods plugin with configurable hooks and trust gate
The ruflo-mods plugin (v0.6.0) is now available, providing a suite of optional, opt-in features for Claude Code. Key capabilities include a mod trust gate to observe or refuse risky newly installed mods, a session budget ladder with a hard stop for subagents, and an agent trimmer to hide unused agent types and save tokens. Additional features include in-process delivery screening for prompt injection and secrets, a compaction carry option to preserve swarm and claim state, and observability tools like a capability probe and session rollup ledger. The plugin also integrates with the shared toast system for notifications and includes a startup self-check to alert users if required marketplace plugins are missing.
plugins/ruflo-mods · high confidence
Introduce ruflo-neural-trader plugin with neural trading, risk-gated pipeline, and benchmarking
The ruflo-neural-trader plugin (v0.3.5) is added, wrapping the neural-trader npm package (^2.8.11) to provide neural trading strategies, backtesting, and portfolio optimization. It introduces four specialized agents (trading-strategist, risk-analyst, market-analyst, backtest-engineer) and implements a structured live-execution pipeline (ADR-126 Phase 5) where the risk-analyst acts as a blocking gate, requiring explicit approval before any broker calls. The plugin includes a tighten-only guard to refuse calls holding secrets, supports dispatching heavy jobs to a Managed Agent cloud runtime, and provides a benchmark suite for backtest throughput, memory recall, and portfolio conjugate-gradient performance.
plugins/ruflo-neural-trader · high confidence
Introduce ruflo-ruvllm plugin with local inference tools and secret guard
The new ruflo-ruvllm plugin (v0.3.4) provides local LLM inference capabilities, including model configuration, MicroLoRA fine-tuning, SONA real-time adaptation, and chat formatting for local model templates. It introduces a tighten-only secret guard that automatically blocks write tool calls containing credentials (keys, tokens, passwords) to prevent them from being trained into adapters or indexed in routing patterns. The plugin includes a /ruvllm-mod command for status, scanning text for secrets, and listing connected tools, along with a status file that tracks blocked calls.
plugins/ruflo-ruvllm · high confidence
Introduce ruflo-x-gateway plugin with Claude directory submission and secret guard mod
This change adds the ruflo-x-gateway plugin (version 0.10.0), which exposes a Streamable HTTP MCP surface at https://x.ruv.io/claude/mcp for coordinating AI-agent swarms via signed Nostr federation, OAuth-protected publishing, and work claims. The plugin includes a complete Claude directory submission bundle (manifest, icon, test cases, and review guide) and a Dockerfile for deployment. It also introduces a local 'mod' hook that screens tool inputs for secrets (keys, tokens, passwords) and refuses to publish them to the shared swarm, with a local slash command (/xgw-mod) to check status and scan text.
plugins/ruflo-x-gateway · high confidence
Introduce v3 MCP server with connection pooling, OAuth 2.1, and rate limiting
The v3/@claude-flow/mcp package now provides a standalone, high-performance Model Context Protocol server implementation. This release adds a configurable connection pool to manage transport connections efficiently, OAuth 2.1 authentication with PKCE support for securing HTTP endpoints, and a token-bucket rate limiter to protect against abuse. It also includes new registries for managing resources, prompts, and async tasks, along with a sampling manager for server-initiated LLM calls and a JSON schema validator for tool inputs.
v3/@claude-flow/mcp · high confidence
Introduce v3 core architecture with type-safe configuration and event-driven orchestration
The shared library now provides the foundational v3 architecture, starting with a new, type-safe configuration system in \src/core/config\ that uses Zod schemas for validation, default presets for agents and swarms, and a loader that merges file, environment variable, and default settings. It also introduces a core event bus (\src/core/event-bus\) supporting synchronous and asynchronous pub/sub with secure ID generation, alongside a comprehensive set of Domain-Driven Design interfaces (\src/core/interfaces\) for agents, tasks, memory, and coordination. These interfaces are wired together by decomposed orchestrator components, including an event coordinator for routing and a health monitor for periodic system checks.
v3/@claude-flow/shared · high confidence
Introduces A2A Agent Card federation and hardened federation policy controls
The plugin now supports discovery and registration of external peers via the Linux Foundation A2A (Agent2Agent) protocol by mapping ruflo federation manifests to A2A Agent Cards and consuming remote cards into the federation registry. This includes a read-only HTTP endpoint at \/.well-known/agent-card.json\ that binds to localhost by default for security. Additionally, the federation layer now enforces stricter authorization policies: inbound messages are validated against a defined set of claims (e.g., \federation:write\, \federation:spawn\) and trust levels, and a circuit breaker service monitors peer spend and failure rates to automatically suspend or evict untrusted or costly peers.
v3/@claude-flow/plugin-agent-federation · high confidence
Introduces comprehensive multi-agent security, governance, and policy analysis primitives
The guidance package now includes a suite of new modules for securing and managing multi-agent systems. The \adversarial\ module provides threat detection, collusion identification, and memory quorum consensus. The \authority\ module enforces human and institutional oversight with cryptographic audit trails for interventions. The \capabilities\ module introduces a typed permission algebra for granular, delegatable, and revocable access control. The \coherence\ module monitors agent behavior drift and enforces economic budgets (tokens, time, cost). The \compiler\ parses CLAUDE.md into structured policy bundles, while the \analyzer\ scores and auto-optimizes these files. Additionally, the \artifacts\ module provides a tamper-evident ledger for production outputs, and the \continue-gate\ prevents runaway agent loops by evaluating step-level context.
v3/@claude-flow/guidance · high confidence
Introduces v3 swarm architecture with agent pooling, CQRS task/agent commands, and pluggable consensus transports
The swarm package now provides a new v3 architecture centered on an AgentPool for lifecycle and auto-scaling, and an application layer using CQRS command handlers (SpawnAgent, TerminateAgent, CreateTask, CancelTask) orchestrated by SwarmApplicationService. Consensus protocols (Raft, Byzantine, Gossip) are unified under ConsensusEngine and now support ADR-095 G2 pluggable transports (LocalTransport and FederationTransport) with optional Ed25519 message signing, enabling secure multi-host consensus wiring. Also in this area: 1 commit (0 fixes) modifying v3/@claude-flow/swarm; Swarm consensus and retry logic improvements; Fix topology wiring for downgraded leader nodes; Fix worker cancellation handling and disclose synthetic analysis results; CLI reliability and security hardening.
v3/@claude-flow/swarm · medium confidence · unverified
Introducing ruRE: Static Reverse-Engineering Plugin for JS, Binaries, .NET, EVM, and CUDA
The new \ruflo-rure\ plugin (v0.6.0) provides a suite of eight read-only static analysis tools that inspect local application artifacts without executing them. It covers five domains: Phase 1 analyzes JavaScript and Electron apps (including \.asar\ unpacking) to map modules and detect process boundaries; Phase 2 inspects native binaries (ELF, PE, Mach-O) for structure and symbols; Phase 3 reads .NET assembly metadata; Phase 4 inventories EVM smart-contract bytecode for hazardous opcodes and proxies; and Phase 5 analyzes CUDA fatbin/cubin/PTX files for kernel resources and generates theoretical occupancy plans. All extracted strings and identifiers are screened by the AIDefence gate, and the tools enforce strict resource caps to prevent abuse.
plugins/ruflo-rure · high confidence
Introducing ruflo-arena: competitive ruliology for program strategies
The new ruflo-arena plugin enables competitive ruliology by letting deterministic finite-state program strategies compete in arenas, tournaments, and adaptive co-evolution simulations. Users can run head-to-head matches, round-robin tournaments that produce competitive arrays and rankings, and hill-climb or mutual co-evolution runs to discover winning strategies empirically. The plugin exposes these capabilities via MCP tools (arena/run, tournament/run, evolve/run, coevolve/run, run/get, run/list) and a local CLI, with all runs being reproducible via seeds and persisted as artifacts for later querying.
plugins/ruflo-arena · high confidence
Introducing ruflo-music plugin for AI music generation
The new ruflo-music plugin enables AI music generation via the Cognitum Music (cogmusic) MCP server. It provides a complete pipeline for composing lyrics, generating tracks, separating stems, extracting MIDI, and mastering audio through a user's own Cognitum account. The plugin includes two agents (music-composer and music-producer), seven skills for end-to-end music workflows, and a local command-line interface. It also features a tighten-only security guard that prevents secrets from being sent to the music service or stored in memory, with a status file and scan command for monitoring.
plugins/ruflo-music · high confidence
Knowledge graph plugin ships with security guard and corrected tool references
The ruflo-knowledge-graph plugin (v0.3.4) is now available as a mod that includes a tighten-only security guard. This guard prevents secrets (keys, tokens, passwords) from being written to the knowledge graph and requires an explicit \confirm: true\ flag for deleting graph nodes or edges. The plugin also corrects a previous bug where skills referenced a non-existent \embeddings\_embed\ tool, switching to the actual \embeddings\_generate\ tool. Additionally, the plugin registers a \/kg-mod\ command for checking status, scanning text for secrets, and viewing recent tool usage.
plugins/ruflo-knowledge-graph · high confidence
MetaHarness plugin introduces security guard and optional integration
The ruflo-metaharness plugin is introduced as an optional integration that surfaces MetaHarness capabilities (score, genome, mcp-scan, threat-model, drift detection) via skills and MCP tools. It enforces ADR-150 by remaining a removable augmentation with graceful degradation when the upstream metaharness CLI is unavailable. A new tighten-only guard (ADR-445) is added to refuse tool calls containing secrets and to require explicit confirmation for flywheel promotion or evidence-reset operations.
plugins/ruflo-metaharness · high confidence
New AI-driven workflow definitions for system testing and hardening
Added three new workflow scripts in .claude/workflows that define automated, agent-driven processes for the platform. The 'full-system-test' workflow orchestrates parallel checks for CLI builds, unit tests, runtime smoke tests, and plugin contracts, synthesizing a single pass/fail report. The 'intelligence-system-hardening' workflow guides agents through implementing specific audit fixes (such as CLI flag parsing and metric fabrication), validating builds, running performance benchmarks, and updating documentation with measured values. The 'plugin-contract-audit' workflow sweeps all plugin smoke contracts, diagnoses failures, and reports a punch list of issues.
.claude/workflows · high confidence
New ASCII-based visual components for boot, data, and hive views
The console now includes a suite of new graphics hooks for rendering ASCII art and terminal-based visualizations. This introduces a cyberpunk-themed boot screen with a constellation map and self-check status, a BBS-style boot sequence with a neon sign, and various data visualization charts including learning pipelines, radar plots, budget gauges, and sparklines. It also adds visualizations for the Hive-Mind voting chambers, the Self-Evolution loop, and the Claims flow, along with supporting utilities for fonts and raster graphics.
plugins/ruflo-console/hooks/gfx · high confidence
New Claude Code configuration with hooks, permissions, and statusline tools
The .claude directory now includes a complete configuration set for the Claude Code CLI. settings.json defines the default model (claude-sonnet-5), enables v3 agent teams and hooks, sets specific bash/MCP permissions, and registers PreToolUse, PostToolUse, and other lifecycle hooks via a hook-handler script. A new mcp.json file configures the 'flow-nexus' MCP server pointing to a Supabase SSE endpoint. Additionally, three new statusline scripts (statusline.sh, statusline.mjs, statusline-command.sh) are added to display real-time metrics like swarm status, memory usage, and task progress in the terminal.
.claude · high confidence
New Codex CLI with dual-mode swarm orchestration and policy-integrated execution
The @claude-flow/codex package now provides a dedicated CLI for OpenAI Codex integration, featuring an init command to scaffold projects with AGENTS.md, SKILL.md, and config.toml templates, alongside a new dual-mode command that orchestrates collaborative swarms of Claude Code and Codex workers. This orchestration supports parallel or chained execution, shared memory via Ruflo, and worktree isolation for concurrent writers. Worker execution is gated by ADR-324 policy preflight checks and ADR-377 caller-identity verification, ensuring that every action is authorized before the worker spawns. The package also includes a harness layer for build evidence and repository state tracking to support reproducible, auditable releases.
v3/@claude-flow/codex · high confidence
New V3 Embedding Service with Multi-Provider Support and Advanced Utilities
The \@claude-flow/embeddings\ package introduces a comprehensive V3 embedding service that supports multiple providers including OpenAI, Transformers.js (with a secure migration to \@huggingface/transformers\), Agentic-flow, and a new lightweight RVF hash-based provider. This update adds document chunking utilities with configurable strategies (sentence, paragraph, character, token), embedding normalization (L2, L1, min-max, z-score), and hyperbolic embedding conversions (Poincaré ball). It also includes a persistent SQLite-backed cache, a pure-TS binary cache for RVF embeddings, and neural substrate integration for semantic drift detection and memory management.
v3/@claude-flow/embeddings · high confidence
New agentic COW memory exploration and local GGUF inference engine
The CLI now includes two new capabilities: a speculative branch-and-promote system for agentic COW memory exploration (agenticow) that allows parallel A/B solution testing with promotion gates, and a pure Node.js GGUF inference engine that provides local model loading and token generation with optional node-llama-cpp integration.
v3/@claude-flow/cli · high confidence
New claims module for issue claiming, work stealing, and load balancing
The \v3/@claude-flow/claims\ module is now available, providing the core infrastructure for federated issue claiming and work coordination. This release introduces a full domain layer with event sourcing (ADR-007) and business rules (ADR-016), supporting both human and agent claimants. Key capabilities include core claiming operations (claim, release, handoff), a work-stealing service for redistributing stale or blocked work, and a load balancer for swarm-wide rebalancing. The module exposes these features via MCP tools (e.g., \claims/issue\_claim\, \claims/swarm\_rebalance\) and CLI commands, enabling agents to manage their workload and coordinate across the swarm.
v3/@claude-flow/claims · high confidence
New console hooks for claims, ADRs, and activity logging
The console now includes dedicated hooks for managing claims (claim, release, handoff, steal), Architecture Decision Records (ADR lifecycle, mission attachment, scope checking), and persistent activity logging (batched event writes, live timeline updates, and alert rule evaluation). These hooks provide the underlying logic for the corresponding UI views and palette commands, ensuring that writes are safe (link-checked, bounded, and retryable) and that model-initiated actions are properly attributed and confirmed.
plugins/ruflo-console/hooks · high confidence
New console views for ADRs, agents, AgentDB, Project Anatole, and Autopilot
The console now includes dedicated views for managing Architecture Decision Records (searchable list, detail with link directions, lint health, and mission attachment), drilling into individual agent state and logs, monitoring the ruflo-agentDB mod's recall and secret-guard status, displaying Project Anatole security rules and alerts, and controlling the Autopilot loop (envelope editing, spend/concurrency gates, parked queue, and start/pause/stop). These views render the underlying data and provide interactive controls for each capability.
plugins/ruflo-console/hooks/views · high confidence
New data layer for Activity, ADR, and Alert management
The console introduces a dedicated data layer in \plugins/ruflo-console/hooks/data\ to power the new Activity, ADR, and Alert views. This includes \activity-store.ts\ for persisting and parsing event logs and lane data, \adr.ts\ and \adr-lint.ts\ for parsing, linking, and validating Architecture Decision Records, \adr-scope.ts\ for checking file changes against ADR scopes, \alerts.ts\ for deriving health alerts and approval queues from state, and \agentdb-mod.ts\ for reading agent database mod status.
plugins/ruflo-console/hooks/data · high confidence
New deployment module for release management and publishing
The v3/@claude-flow/deployment package introduces a new suite of tools for managing software releases. It includes a ReleaseManager for handling version bumping (major, minor, patch, prerelease), changelog generation, and git tagging, along with a Publisher class to handle npm package publishing with support for tags, access levels, and dry runs. A Validator component ensures release readiness by checking package.json structure, git status, dependencies, and running lint/test/build commands. The module also provides legacy compatibility functions (prepare, deploy) for existing workflows.
v3/@claude-flow/deployment · high confidence
New development helper suite and safety mechanisms for .claude/helpers
The .claude/helpers directory now contains a comprehensive set of new utility scripts and hooks to support the V3 development workflow. This includes an ADR compliance checker (adr-compliance.sh) to verify architectural decisions, a DDD progress tracker (ddd-tracker.sh), and a checkpoint manager (checkpoint-manager.sh) for Git-based session rollbacks. Operational safety is enhanced with a daemon manager (daemon-manager.sh) for background services, an auto-commit helper (auto-commit.sh), and a context persistence hook (context-persistence-hook.mjs) that archives conversation history to SQLite or JSON to prevent data loss during compaction. Security hardening is introduced via github-safe.js, which prevents shell injection in GitHub CLI commands by using temporary files for body content, and aggressive-microcompact.mjs, which optimizes context window usage. Additionally, auto-memory-hook.mjs bridges session data to the AgentDB, and a new .LOCKED file protects the directory from accidental overwrites during development.
.claude/helpers · high confidence
New example plugins for plugin creation and RuVector-powered AI capabilities
This update adds two new example plugin collections to the \@claude-flow/plugins/examples\ package. The \plugin-creator\ plugin introduces a meta-plugin that allows users to programmatically generate new plugins with various configurations, including tools, hooks, workers, and swarm integration, complete with code generation and validation. Additionally, a new suite of six RuVector plugins is introduced, leveraging \@ruvector/wasm\ for high-performance vector operations. These include a Reasoning Bank for storing and retrieving similar reasoning trajectories, Semantic Code Search for natural language code queries, a SONA Learning plugin for adaptive learning patterns, an Intent Router for smart query-to-agent mapping, an MCP Tool Optimizer for suggesting optimal tool sequences, and a Hook Pattern Library for learning and recommending effective hook configurations based on file types and operations.
v3/@claude-flow/plugins/examples · high confidence
New performance verification system with per-OS capability baselines
The verification folder now includes a performance layer that tracks key user-visible operations (such as package installation, memory module loading, and witness verification speed) across macOS, Linux, and Windows. This system appends timing data to OS-specific JSONL files, allowing developers to catch regressions by comparing current runtimes against rolling baselines. It complements the existing cryptographic witness system, which attests to fix presence, by ensuring that fixes do not inadvertently slow down critical paths.
verification · high confidence
New reinforcement learning algorithms: A2C and Curiosity-Driven Exploration
The neural module now includes two new reinforcement learning algorithms: Advantage Actor-Critic (A2C) and Curiosity-Driven Exploration. The A2C implementation provides a synchronous actor-critic network with N-step returns, entropy regularization, and advantage normalization, targeting less than 10ms per update step. The Curiosity module implements intrinsic motivation for exploration using both Intrinsic Curiosity Module (ICM) and Random Network Distillation (RND) approaches, with forward and inverse dynamics models to generate exploration bonuses, targeting less than 5ms per forward pass.
v3/@claude-flow/neural · high confidence
New ruflo-deepseek-harness plugin with secret guard mod
The new ruflo-deepseek-harness plugin (v0.2.4) exposes DeepSeek's chat and reasoning models as ruflo skills, providing \deepseek-chat\ for single-turn completions and \deepseek-reason\ for reasoning-mode completions that surface the model's chain-of-thought. As a function-hook mod, it includes a tighten-only secret guard that refuses Bash commands invoking the plugin's scripts if the prompt contains secrets (keys, tokens, or passwords) destined for the DeepSeek API. The mod also registers a local \/deepseek-mod\ command for status and scanning, and writes a session status file at \.claude-flow/deepseek-mod/status.json\. The plugin is designed for graceful degradation, exiting with a degraded status if the API key is missing or the service is unreachable.
plugins/ruflo-deepseek-harness · high confidence
New ruflo-docs mod with secret-scanning guard and /docs-mod CLI
The ruflo-docs plugin now includes a new mod that enforces a secret-scanning guard on documentation-related tool calls (memory\_store and hooks\_worker-dispatch) to prevent secrets from being written to shared documentation. The mod registers a local /docs-mod command allowing users to check status, scan text for potential secrets, and list documentation files. A status file is written to track blocked calls and guard state.
plugins/ruflo-docs/hooks · high confidence
New ruflo-ruos plugin for running swarm agents on ruOS cloud desktops
The ruflo-ruos plugin introduces the ability to place ruflo swarm agents on your own ruOS cloud desktops, stream their output back into swarm state, and hand finished work back for review. It supports two transports: the tenant-authenticated ruOS fleet MCP/REST (the default, required for external access) and per-tenant SSH on port 2222 (desktop-to-desktop only). The plugin provides commands to list hosts, run agents with configurable prompts and timeouts, view desktops, and perform read-only deploy hand-offs. It includes a mod that adds a 'ruos' segment to ruflo's status line, showing agent counts and desktop names by reading a local snapshot file. The plugin enforces security by never contacting the unauthenticated desktop executor port, never issuing deletion commands, and using audited command builders with nonce markers to prevent echo-based injection. It also respects ruOS's weekday 23:00 America/Toronto auto-stop, refusing runs that would cross this window unless explicitly ignored.
plugins/ruflo-ruos · high confidence
New unified plugin SDK with collection management and advanced vector integrations
The @claude-flow/plugins package introduces a comprehensive plugin development framework for v3, featuring a core plugin interface and builder for creating agents, MCP tools, hooks, and workers. It adds a PluginCollectionManager to group, activate, and persist plugin sets, along with several official collections (core, development, intelligence, swarm, security, utility). The SDK also integrates with external systems via an AgenticFlowBridge for swarm coordination and a RuVector PostgreSQL Bridge that exposes vector search, attention mechanisms, GNN layers, and hyperbolic embeddings as MCP tools.
v3/@claude-flow/plugins/src · high confidence
New v3 security module with CVE remediation, agent authorization, and input validation
The v3/@claude-flow/security module introduces a comprehensive security layer for the platform. It addresses critical vulnerabilities (CVE-2, CVE-3, HIGH-1, HIGH-2) by replacing weak password hashing with bcrypt, removing hardcoded credentials in favor of secure generation, preventing command injection via a safe executor, and validating file paths to stop traversal attacks. The module also implements ADR-144 P1 for agent authorization propagation, ensuring that delegated tasks respect scope boundaries and cannot escalate privileges. Additionally, it provides robust input validation using Zod schemas, threat detection for SQL injection and XSS, and an MCP composition inspector to detect malicious tool descriptions. A new security application service orchestrates these checks, allowing for context-aware validation of paths, commands, and user inputs.
v3/@claude-flow/security · high confidence
Security audit plugin with secret guard and mod integration
The ruflo-security-audit plugin (v0.3.4) now includes a function-hook mod that prevents live secrets from being stored in security-related memory namespaces (security, audit, cve, vuln, findings, secaudit). The guard, enabled by default via the userConfig option, intercepts tool calls to memory\_store and agentdb stores, screening values for secret patterns (API keys, tokens, passwords) and refusing writes that contain them. A local slash command /secaudit-mod provides status, scanning, and namespace information. The plugin also aligns its CLI usage (e.g., --depth deep instead of full) and corrects command descriptions in its skills and agent definitions.
plugins/ruflo-security-audit · high confidence
Standalone terminal (ruflo tui) with AI page and security guards
The \ruflo tui\ command now runs the console as a standalone terminal interface, introducing a new AI page that lets users ask Codex (using a strictly pinned, read-only version) or OpenRouter. To ensure safety, the implementation includes an \ai-guard\ module that validates all AI requests against a strict allow-list of hosts and flags, and a \parity\ map that defines which console views are supported in this standalone mode. The terminal also features a new ANSI renderer for accurate display of wide characters and colors, and a confirmation system for actions.
plugins/ruflo-console/tui · high confidence
V3 Hooks system introduces pattern learning, guidance, and official bridge
The @claude-flow/hooks package now includes a complete V3 implementation featuring a ReasoningBank for vector-based pattern learning (using HNSW indexing and AgentDB persistence), a GuidanceProvider that generates Claude-visible context and pre-edit security checks (blocking sensitive files like .env), and an OfficialHooksBridge that maps internal V3 hook events to the standard Claude Code plugin API. This also adds a CLI for querying guidance and a daemon manager for background metrics and swarm monitoring.
v3/@claude-flow/hooks · high confidence
V3 introduces structured agent definitions and AIDefence security documentation
The v3 workspace now includes a set of YAML configuration files for core agent roles (architect, coder, reviewer, security-architect, tester) under \@claude-flow/agents\, each declaring version 3.0.0, specific capabilities, and optimization strategies. Additionally, an \intelligence.json\ file has been added to track routing metrics and command patterns, and the \@claude-flow/aidefence\ package now ships with comprehensive documentation detailing its AI Manipulation Defense System capabilities, including threat detection, PII scanning, and self-learning features. Also in this area: Fix Zod schema definitions for record types to enforce value constraints.
v3, v3/@claude-flow/browser · high confidence
ruflo-agent plugin introduces cloud-managed agent runtime and nested sub-agent orchestration
The ruflo-agent plugin now supports two agent runtimes: the existing local WASM-sandboxed agents and a new Anthropic Claude Managed Agents cloud backend, accessible via \managed\agent\\*\ MCP tools. It also adds infrastructure for nested sub-agent trees (up to depth 5) with specialized orchestrator and leaf agent templates (e.g., \nested-queen\, \nested-researcher\) that integrate with ruflo's intelligence pipeline, authorization scope reduction, and content-gating (AIDefence) mechanisms. The plugin's secret-screening guard has been refined to judge values rather than names, reducing false positives, and the plugin version is updated to 0.3.3.
plugins/ruflo-agent · high confidence
ruflo-browser plugin introduces session-as-skill architecture with tighten-only security guard
The ruflo-browser plugin (v0.3.5) adopts a session-as-skill architecture where every browser session is captured as a first-class RVF cognitive container with a ruvector trajectory, indexed in AgentDB, and gated by AIDefence for PII and prompt-injection safety. It exposes 23 Playwright-backed MCP tools (18 interaction primitives plus 5 lifecycle tools for recording, ending, replaying, template application, and cookie use) and a \/ruflo-browser\ slash command for session management. As a mod, it installs a tighten-only tool guard that refuses browser\_open calls to file/javascript/data/chrome URLs, URLs with embedded credentials, cloud metadata hosts, and unsafe Chrome launch flags, as well as browser\_eval scripts that read cookies or storage and can exfiltrate them. The guard is configurable via \userConfig\ options \guard\ (on/off, default on) and \strictUrls\ (on/off, default off), and provides a local \/browser-mod\ command for status, script scanning, and URL checking without model turns. Also in this area: Fix browser doctor path for unrelated working directories.
plugins/ruflo-browser/scripts · high confidence
ruflo-ddd plugin adds secret guard and /ddd-mod command
The ruflo-ddd plugin now includes a native secret guard that blocks domain-model writes (via memory\_store and agentdb\_hierarchical-store) if they contain secrets such as keys, tokens, or passwords, and provides a /ddd-mod command to check guard status, scan text for secrets, and list bounded contexts. A status file at .claude-flow/ddd-mod/status.json tracks blocked calls and guard state.
plugins/ruflo-ddd/hooks · high confidence
ruflo-jujutsu plugin introduces diff analysis and a tighten-only security guard
The ruflo-jujutsu plugin (v0.3.6) provides advanced git workflows by wrapping six analyze\\ MCP tools (diff, diff-risk, diff-classify, diff-reviewers, file-risk, diff-stats) to offer risk scoring, change classification, and reviewer role suggestions based on file paths and risk rather than history. It includes a /jujutsu command and a git-specialist agent. Additionally, as a mod (ADR-445), it registers a tighten-only guard that refuses PR create/review calls containing secrets in the text and requires explicit confirmation for PR merge/close actions, while providing a /jujutsu-mod command to check status, scan text, and view recent tool usage.
plugins/ruflo-jujutsu · high confidence
ruflo-loop-workers plugin introduces background automation with security guards
The ruflo-loop-workers plugin (v0.3.4) provides cache-aware /loop workers and CronCreate background automation, wrapping five hooks\worker-\ MCP tools and exposing 12 background worker triggers (ultralearn, optimize, consolidate, predict, audit, map, preload, deepdive, document, refactor, benchmark, testgaps). As a mod, it enforces a tighten-only guard on worker dispatch to prevent secrets from entering the worker context and limits per-session dispatches (default 100, configurable up to 1000), while providing a /loop-mod command to inspect status, scan text, and view recent activity.
plugins/ruflo-loop-workers · high confidence
ruflo-rag-memory plugin introduces secret guard and mod infrastructure
The ruflo-rag-memory plugin now includes a 'mod' (ADR-445) that acts as a tighten-only guard for its write tools, preventing secrets (keys, tokens, passwords) from being stored in memory. This is controlled by a new 'guard' userConfig option (default 'on') and enforced via a shared screening logic in hooks/screen.ts. The plugin also registers a '/rag-mod' command for status, scanning, and tool listing, and maintains a session status file at .claude-flow/rag-mod/status.json. Additionally, the plugin's metadata (plugin.json) has been updated to version 0.3.5 with new keywords (mcp, claude-memories, bridged-memory) and documentation (README, EXAMPLES, CHANGELOG) has been added or updated to reflect these changes and correct previous command inaccuracies.
plugins/ruflo-rag-memory · high confidence
ruflo-sparc v0.3.4 adds native Codex support and a secret-detection guard
The ruflo-sparc plugin (v0.3.4) now works in both Claude and Codex. For Codex, a new .codex-plugin/plugin.json and hooks/codex-hooks.json wire up a native PreToolUse adapter (hooks/codex.ts → hooks/codex-hook.cjs) that screens writes to sparc memory namespaces for secrets using the shared screen logic (hooks/screen.ts, hooks/guard.ts) and exposes a local /sparc-mod command (hooks/command.ts) for status, scanning, and phase info. A Claude-side mod (hooks/hooks.json → hooks/register.ts) provides the same secret guard, status file, and /sparc-mod command for Claude sessions. The plugin.json declares a userConfig guard option (on/off, default on) to tighten-only refuse specs or phase artifacts containing secrets, and the README documents the five-phase SPARC workflow, sibling-plugin alignments, and namespace coordination.
plugins/ruflo-sparc · high confidence
ruflo-swarm plugin v0.3.7: live swarm pane, multi-agent coordination, and ADR injection
The ruflo-swarm plugin (v0.3.7) introduces a live swarm pane that visualizes agent tiles, topology, tasks, and hive-mind votes, with buttons to stop agents, claim tasks, and vote. It wraps 12 MCP tools (4 swarm\\, 8 agent\\) and supports 6 topologies (hierarchical, mesh, hierarchical-mesh, ring, star, adaptive). A new feature injects accepted ADR digests from ruflo-console missions into spawned subagent prompts (configurable via injectAdrs). The plugin also adds a shared toast system, an audit trail, and a Monitor-friendly poller for swarm status changes.
plugins/ruflo-swarm · high confidence
ruflo-testgen plugin introduces Test-Driven Repair and session-level hook mod
The ruflo-testgen plugin (v0.3.3) adds a Test-Driven Repair capability that drives a headless \claude -p\ agent to fix failing tests by editing source code while strictly preserving the test file, enforcing a $5 budget cap and Read/Edit/Bash-only tool access. It also introduces a session-level hook mod that monitors and counts testgen-related tool calls (like \hooks\_worker-dispatch\ and coverage commands) via an observe-only guard, writing session statistics to a local status file and exposing them through the \/testgen-mod\ command.
plugins/ruflo-testgen · high confidence
ruflo-workflows plugin introduces dual workflow surfaces and GAIA benchmark integration
The ruflo-workflows plugin now exposes two complementary workflow automation surfaces: a persisted, state-machine-driven MCP interface (10 tools for create, run, pause, resume, cancel, etc.) and a native Claude Code JavaScript orchestration surface for deterministic subagent fan-out via \.claude/workflows/\*.js\ scripts. It also adds a comprehensive GAIA benchmark suite, including commands to run benchmarks (\/gaia run\), view history and costs, check the HAL leaderboard, and submit signed, audited results. A new function-hook mod provides a configurable secret guard to refuse workflow definitions or runs containing sensitive data.
plugins/ruflo-workflows · high confidence
Security
Security guard and diagnostic commands for BBS and ChatGPT federation plugins
The ruflo-bbs-federation and ruflo-chatgpt-federation plugins now include a tighten-only security guard that intercepts their federation tool calls to prevent data leaks. The guard blocks messages containing detected secrets (such as API keys, tokens, and passwords) or credential-bearing fields, refuses peer URLs that include embedded credentials, and restricts ChatGPT publishing to public channels with bounded payload sizes. A wildcard bind option is also guarded for the BBS plugin. To help users verify safety before sending, new local diagnostic commands (/bbs-mod and /chatgpt-mod) allow scanning text and checking peer/channel validity without consuming a model turn, and session status files are written to report guard activity.
plugins/ruflo-bbs-federation/hooks, plugins/ruflo-chatgpt-federation/hooks · high confidence
Behavioural changes
11 commits (4 fixes) modifying plugins/ruflo-protector
A change to existing behaviour in plugins/ruflo-protector — 11 commits (4 fixs), 25 files: “fix(mods): a toast needs no engine.create (bind at session.start too), a throwing clock falls back to the wall clock; update the cost ladder expectation in the v3 mods contract”; “feat(ruflo-protector): Project Anatole plugin, tests, corpus gate, bench, fleet registration”; “chore(plugins): give every plugin manifest a repository field”; and 8 more (refs \#4202, \#4211, \#4096).
plugins/ruflo-protector · medium confidence · unverified
2 commits (1 fix) modifying plugins/ruflo-agntcy/src
A change to existing behaviour in plugins/ruflo-agntcy/src — 2 commits (1 fix), 9 files: “ADR-378/379/380: npm Trusted Publishing, statusline segments, AGNTCY/Outshift runtime integration (\#2879)”; “fix(mods): vitest suites in agntcy, arena and graph-intelligence are \*.spec.ts so claude plugin test collects only the kit tests” (refs \#2879).
plugins/ruflo-agntcy/src · medium confidence · unverified
28 commits (11 fixes) modifying plugins/ruflo-graph-intelligence
A change to existing behaviour in plugins/ruflo-graph-intelligence — 28 commits (11 fixs), 49 files: “fix(mods): secret screen judges values, not names — fewer false positives, vendor keys, URL credentials, env assignments, 200 KB scan (ADR-445)”; “feat(graph-intelligence): \#2044 ADR-123 Phase 1 — ruflo-graph-intelligence plugin foundation”; “feat(plugins): goals, graph-intelligence, intelligence, iot-cognitum as mods”; and 25 more (refs \#2044, \#2, \#4442, \#4444, \#4202, …).
plugins/ruflo-graph-intelligence · medium confidence · unverified
32 commits (14 fixes) modifying plugins/ruflo-goals
A change to existing behaviour in plugins/ruflo-goals — 32 commits (14 fixs), 30 files: “darwin-plugins iter 1 — ruflo-goals: use memory\_list (not memory\search '\') in /goals”; “docs(adr): bulk-update plugin contract ADR statuses (sweep)”; “fix(mods): secret screen judges values, not names — fewer false positives, vendor keys, URL credentials, env assignments, 200 KB scan (ADR-445)”; and 29 more (refs \#1726, \#2, \#4202, \#4211, \#3558, …).
plugins/ruflo-goals · medium confidence · unverified
352 commits (128 fixes) modifying scripts
A change to existing behaviour in scripts — 352 commits (128 fixs), 177 files: “Checkpoint: File edits”; “test(mods): check that every tool name a plugin guard watches exists in the registry”; “fix(plugins): drop unregistered optionalDependencies; guard monteCarlo on empty input”; and 303 more (refs \#2046, \#4, \#3921, \#2267, \#2257, …).
scripts · medium confidence · unverified
4 commits (2 fixes) modifying (repo-wide)
A change to existing behaviour in (repo-wide) — 4 commits (2 fixs), 83 files: “fix: move @claude-flow/shared to its own PR; codex VERSION 3.0.7; floor tests survive the root ratchet”; “fix(zod4): port shared and v3 plugin schemas to zod 4; gastown convoy IDs from crypto.randomUUID”; “feat(plugins): Add 10 RuVector WASM plugins with 50 MCP tools (\#1000)”; and 1 more (refs \#1000).
(repo-wide) · medium confidence · unverified
53 commits (24 fixes) modifying plugins/ruflo-cost-tracker/scripts
A change to existing behaviour in plugins/ruflo-cost-tracker/scripts — 53 commits (24 fixs), 45 files: “fix(cost-tracker): log Stop-hook telemetry write failures instead of discarding them (\#3227)”; “feat(cost-tracker): wire codemod benchmark into cost-trend (ADR-143 follow-up)”; “feat(cost-tracker): P11 — cost-per-conversation view (v0.12.0)”; and 49 more (refs \#3227, \#1902, \#2132, \#3558, \#4490, …).
plugins/ruflo-cost-tracker/scripts · medium confidence · unverified
6 commits (1 fix) modifying v3/@claude-flow/providers
A change to existing behaviour in v3/@claude-flow/providers — 6 commits (1 fix), 14 files: “Replace the models api.anthropic.com retired (\#4512)”; “feat(cli,codex): ecosystem integrations + skills; darwin ^0.13.0, ruvllm ^4.1.0; release 3.72.1 (\#4838)”; “checkpoint: File edit:”; and 3 more (refs \#4512, \#4838, \#2237, \#4673).
v3/@claude-flow/providers · medium confidence · unverified
7 commits (3 fixes) modifying plugins/ruflo-agntcy/scripts
A change to existing behaviour in plugins/ruflo-agntcy/scripts — 7 commits (3 fixs), 1 file: “feat(mods): vendor the flag option parser from one origin with the screen sync and drift check (review \#2)”; “feat(ruflo-agntcy): mod (guard/status/command) as function hooks, ADR-445 pattern”; “feat(plugins): cut per-turn footprint — user-invoked operational skills, description ceiling, footprint report (ADR-507) (\#4588)”; and 4 more (refs \#2, \#4588, \#4096).
plugins/ruflo-agntcy/scripts · medium confidence · unverified
96 commits (41 fixes) modifying ruflo
A change to existing behaviour in ruflo — 96 commits (41 fixs), 564 files: “release: 3.10.33 + CI regression guards for \#2267 / \#2257 / \#2256”; “release: 3.73.0”; “fix+perf+docs(intelligence): self-learning audit, hardening fixes, honest perf numbers (3.10.7) (\#2228)”; and 91 more (refs \#2267, \#2257, \#2256, \#2228, \#2114, …).
ruflo · medium confidence · unverified
CLI entry points updated for v3 proxying and legacy compatibility
The bin directory now uses new entry-point scripts to support the v3 architecture. bin/cli.js and bin/npx-safe-launch.js have been rewritten to dynamically proxy execution to the v3 CLI implementation (@claude-flow/cli), ensuring cross-platform compatibility. Additionally, bin/npx-repair.js has been introduced as a no-op module to maintain backwards compatibility with older versions of the CLI that may still attempt to import these cache-repair functions.
bin · high confidence
IoT Cognitum plugin adds mod-based security guards and local management commands
The ruflo-iot-cognitum plugin now ships as a mod that enforces security policies on its own tools by default. It includes a write guard that blocks storing secrets (tokens, keys, passwords) in device records and prevents passing secrets on the command line, as well as a destructive-confirmation guard that requires explicit confirmation (via --confirm or the COGNITUM\_IOT\_CONFIRM=1 environment variable) for fleet deletions and device removals. Additionally, a new /iot-mod command provides local status reporting, text scanning for secrets, and device listing without requiring a model turn.
plugins/ruflo-iot-cognitum · high confidence
New repository ignore and publish configurations for v3
The repository now enforces stricter file handling for the v3 architecture. The \.gitignore\ has been completely rewritten to exclude v3 runtime artifacts (such as helper manifests, proven-config state, and compile caches) and specific memory/coordination data, while explicitly preserving plugin-shipped \.mcp.json\ templates and README files. Additionally, a new \.npmignore\ file has been added to ensure that published packages exclude source maps, test files, development configurations, and documentation, significantly reducing package size.
(repo-wide) · high confidence
Plugin directories now point to .claude configuration
The plugin's agents, commands, and skills directories have been converted into symbolic links that point to the corresponding folders within the .claude directory. This change reorganizes where the plugin loads its agent definitions, command structures, and skill files from, ensuring they are sourced directly from the .claude configuration location.
plugin · high confidence
RVF plugin adds secret-detection guard for memory and session writes
The ruflo-rvf plugin now includes a tighten-only security guard that screens write operations (such as memory\_store, session\_save, and hooks\_transfer) for secrets like API keys, tokens, and passwords. When a secret is detected, the call is denied with a message that never reveals the secret or the specific namespace if it belongs to another plugin. Users can monitor the guard's activity via the /rvf-mod command (status, scan, tools) and a status file at .claude-flow/rvf-mod/status.json. The guard is enabled by default but can be disabled via the userConfig option.
plugins/ruflo-rvf · high confidence
Vendored RufloSnapshot type removes legacy policy mode
The \plugins/ruflo-console/types/index.d.ts\ file now defines a vendored subset of the \$.ruflo\ contract from \ruflo-mods\. The \RufloSnapshot\ type has been updated to remove the 'legacy' option from the \policy\ field, which is now restricted to 'none', 'observe', 'enforce', or 'unreadable'. This change ensures the console's type definitions remain synchronized with the underlying mod's policy enforcement logic.
plugins/ruflo-console/types · high confidence
ruflo-plugin-creator v0.4.3: governed mod scaffolding and stricter validation
The ruflo-plugin-creator plugin now scaffolds governed Claude Code mods (function hooks) via the \create-mod\ skill, using a hybrid template that includes a classic fallback and a status file contract for the console's Mods section. The plugin's own \/creator-mod\ command provides local, read-only status, reserved-name checks, and manifest verification. Validation has been tightened: \plugin.json\ must no longer contain \skills\, \commands\, or \agents\ arrays, and the \validate-plugin\ skill now verifies that referenced MCP tools actually exist. The scaffolder also warns against known MCP-tool drifts (e.g., \embeddings\_embed\ vs \embeddings\_generate\, \namespace\ vs \tier\ in AgentDB tools).
plugins/ruflo-plugin-creator · high confidence
Fixes
1 commit (1 fix) fixing v3/@claude-flow/guidance/wasm-kernel
A fix in v3/@claude-flow/guidance/wasm-kernel — 1 commit (1 fix), 1 file: “fix(guidance): scan\_secrets no longer panics on non-ASCII secret values”. Also in this area: V3 introduces structured agent definitions and AIDefence security documentation.
v3/@claude-flow/guidance/wasm-kernel · medium confidence · unverified
10 commits (6 fixes) fixing plugins/ruflo-agntcy/hooks
A fix in plugins/ruflo-agntcy/hooks — 10 commits (6 fixs), 6 files: “fix(mods): secret screen judges values, not names — fewer false positives, vendor keys, URL credentials, env assignments, 200 KB scan (ADR-445)”; “feat(mods): vendor the flag option parser from one origin with the screen sync and drift check (review \#2)”; “fix(mods): one iterative textsOf in the shared screen — depth, sibling, size caps, object keys and the invisible-character set fixed at the root (ADR-445)”; and 7 more (refs \#2, \#4202, \#4211, \#4096).
plugins/ruflo-agntcy/hooks · medium confidence · unverified
11 commits (10 fixes) fixing plugins/ruflo-adr/scripts/lib
A fix in plugins/ruflo-adr/scripts/lib — 11 commits (10 fixs), 3 files: “fix(ruflo-adr): read the leading status keyword, skip index/template files, report Unknown in adr-verify (\#4126)”; “fix(plugins): run the installed ruflo CLI before \npx @claude-flow/cli@latest\ in cost-tracker, adr and metaharness (\#3558)”; “fix: complete reports and consistent initialization for v3.32.37 (\#2851)”; and 8 more (refs \#4126, \#3558, \#2851, \#2788, \#3633, …).
plugins/ruflo-adr/scripts/lib · medium confidence · unverified
15 commits (10 fixes) fixing plugins/ruflo-adr/hooks
A fix in plugins/ruflo-adr/hooks — 15 commits (10 fixs), 10 files: “feat(ruflo-adr): mod (guard/status/command) as function hooks, ADR-445 pattern”; “fix(mods): secret screen judges values, not names — fewer false positives, vendor keys, URL credentials, env assignments, 200 KB scan (ADR-445)”; “feat(mods): vendor the flag option parser from one origin with the screen sync and drift check (review \#2)”; and 12 more (refs \#2, \#4277, \#4202, \#4211, \#4096).
plugins/ruflo-adr/hooks · medium confidence · unverified
17 commits (11 fixes) fixing plugins/ruflo-agentdb/hooks
A fix in plugins/ruflo-agentdb/hooks — 17 commits (11 fixs), 12 files: “fix(mods): secret screen judges values, not names — fewer false positives, vendor keys, URL credentials, env assignments, 200 KB scan (ADR-445)”; “feat(ruflo-agentdb): AgentDB as a mod (ADR-445)”; “feat(mods): vendor the flag option parser from one origin with the screen sync and drift check (review \#2)”; and 14 more (refs \#2, \#4202, \#4211, \#4096).
plugins/ruflo-agentdb/hooks · medium confidence · unverified
19 commits (10 fixes) fixing plugins/ruflo-agentdb/scripts
A fix in plugins/ruflo-agentdb/scripts — 19 commits (10 fixs), 2 files: “fix(plugins): agentdb\_context-synthesize guidance in agentdb, observability and rag-memory (\#4691 follow-up) (\#4705)”; “fix(mods): secret screen judges values, not names — fewer false positives, vendor keys, URL credentials, env assignments, 200 KB scan (ADR-445)”; “feat(ruflo-agentdb): AgentDB as a mod (ADR-445)”; and 16 more (refs \#4691, \#4705, \#4133, \#2, \#4843, …).
plugins/ruflo-agentdb/scripts · medium confidence · unverified
26 commits (15 fixes) fixing plugins/ruflo-adr/scripts
A fix in plugins/ruflo-adr/scripts — 26 commits (15 fixs), 4 files: “feat(ruflo-adr): mod (guard/status/command) as function hooks, ADR-445 pattern”; “fix(ruflo-adr): read the leading status keyword, skip index/template files, report Unknown in adr-verify (\#4126)”; “fix(ci): repair CI regressions surfaced by the integrated community fixes”; and 23 more (refs \#4126, \#3558, \#2, \#2851, \#2788, …).
plugins/ruflo-adr/scripts · medium confidence · unverified
27 commits (14 fixes) fixing plugins/ruflo-ruvector
A fix in plugins/ruflo-ruvector — 27 commits (14 fixs), 22 files: “fix(mods): secret screen judges values, not names — fewer false positives, vendor keys, URL credentials, env assignments, 200 KB scan (ADR-445)”; “fix(mods): a domain plugin's refusal of another plugin's memory\_store no longer claims the write (variant B, review \#8)”; “fix(plugins): keep history in comments, hint on an uncached pin, scope the verified claim”; and 24 more (refs \#8, \#2, \#4855, \#4159, \#4588, …).
plugins/ruflo-ruvector · medium confidence · unverified
32 commits (19 fixes) fixing plugins/ruflo-ai-team
A fix in plugins/ruflo-ai-team — 32 commits (19 fixs), 86 files: “fix(mods): secret screen judges values, not names — fewer false positives, vendor keys, URL credentials, env assignments, 200 KB scan (ADR-445)”; “fix(ai-team): include task history in run evidence”; “fix(ai-team): enlarge workspace rail and restore evidence view (\#3550)”; and 29 more (refs \#3550, \#2, \#3556, \#3549, \#4202, …).
plugins/ruflo-ai-team · medium confidence · unverified
5 commits (3 fixes) fixing v3/plugins/financial-risk
A fix in v3/plugins/financial-risk — 5 commits (3 fixs), 13 files: “fix(plugins): drop unregistered optionalDependencies; guard monteCarlo on empty input”; “fix: move @claude-flow/shared to its own PR; codex VERSION 3.0.7; floor tests survive the root ratchet”; “fix(zod4): port shared and v3 plugin schemas to zod 4; gastown convoy IDs from crypto.randomUUID”; and 2 more (refs \#1000).
v3/plugins/financial-risk · medium confidence · unverified
5 commits (3 fixes) fixing v3/plugins/gastown-bridge
A fix in v3/plugins/gastown-bridge — 5 commits (3 fixs), 62 files: “fix: move @claude-flow/shared to its own PR; codex VERSION 3.0.7; floor tests survive the root ratchet”; “fix(zod4): port shared and v3 plugin schemas to zod 4; gastown convoy IDs from crypto.randomUUID”; “fix(plugins): \#1902/\#1903/\#1904 install-safety + plugin-package CI guard (\#1925)”; and 2 more (refs \#1902, \#1925, \#1002).
v3/plugins/gastown-bridge · medium confidence · unverified
5 commits (3 fixes) fixing v3/plugins/hyperbolic-reasoning
A fix in v3/plugins/hyperbolic-reasoning — 5 commits (3 fixs), 16 files: “fix: move @claude-flow/shared to its own PR; codex VERSION 3.0.7; floor tests survive the root ratchet”; “fix(zod4): port shared and v3 plugin schemas to zod 4; gastown convoy IDs from crypto.randomUUID”; “fix(ruflo-mods): classify risky host calls and intercepting hooks (\#3802)”; and 2 more (refs \#3802, \#1000).
v3/plugins/hyperbolic-reasoning · medium confidence · unverified
5 commits (3 fixes) fixing v3/plugins/prime-radiant
A fix in v3/plugins/prime-radiant — 5 commits (3 fixs), 33 files: “fix: move @claude-flow/shared to its own PR; codex VERSION 3.0.7; floor tests survive the root ratchet”; “fix(zod4): port shared and v3 plugin schemas to zod 4; gastown convoy IDs from crypto.randomUUID”; “fix(ci): prime-radiant TS2307 install-safety failure — indirect the optional-wasm import (\#2586)”; and 2 more (refs \#2586, \#996).
v3/plugins/prime-radiant · medium confidence · unverified
6 commits (4 fixes) fixing v3/plugins/code-intelligence
A fix in v3/plugins/code-intelligence — 6 commits (4 fixs), 13 files: “fix: move @claude-flow/shared to its own PR; codex VERSION 3.0.7; floor tests survive the root ratchet”; “fix(zod4): port shared and v3 plugin schemas to zod 4; gastown convoy IDs from crypto.randomUUID”; “fix(plugin-code-intelligence@3.0.0-alpha.3): real import edges in GNN bridge”; and 3 more (refs \#1000).
v3/plugins/code-intelligence · medium confidence · unverified
6 commits (4 fixes) fixing v3/plugins/healthcare-clinical
A fix in v3/plugins/healthcare-clinical — 6 commits (4 fixs), 13 files: “fix: move @claude-flow/shared to its own PR; codex VERSION 3.0.7; floor tests survive the root ratchet”; “fix(zod4): port shared and v3 plugin schemas to zod 4; gastown convoy IDs from crypto.randomUUID”; “fix: v3.5.43 — resolve 9 issues, replace 22 stubs, fix healthcare tests (\#1435)”; and 3 more (refs \#1435, \#1000).
v3/plugins/healthcare-clinical · medium confidence · unverified
Pre-commit hook blocks commits containing API keys
A new pre-commit hook has been added to automatically scan staged changes for sensitive data like API keys before allowing a commit. The hook executes a redaction validator script; if sensitive data is detected, the commit is blocked with an error message instructing the user to remove secrets. If the validator script is not yet built, the check is skipped with a reminder to run the build process.
.githooks · high confidence
Windows hook stability and MCP launch reliability
The ruflo-core plugin now uses a native Node.js shim (ruflo-hook.cjs) instead of a bash wrapper, ensuring PreToolUse and PostToolUse hooks run reliably on Windows without crashing. This shim includes argument escaping to prevent shell injection risks and ensures hooks always exit cleanly. Additionally, a new launcher script (mcp-launch.cjs) intelligently resolves the local CLI binary before falling back to npx, preventing version drift and ensuring the correct MCP server is started. Also in this area: 3 commits (1 fix) modifying plugins/ruflo-core; Fixes to researcher agent parameters and command recording switch; Fix ruflo CLI resolution to avoid project directory interference.
plugins/ruflo-core/scripts · medium confidence · unverified
Windows-compatible plugin hook shims
Added a Node.js-based shim (ruflo-hook.cjs) to replace the shell-based hook invoker on Windows, resolving the 'cannot execute binary file' error that previously blocked hook execution on native Windows environments. The new shim mirrors the behavior of the existing shell script (ruflo-hook.sh) by reading hook payloads from stdin, preferring locally installed binaries over npx, and ensuring hooks always exit successfully to avoid blocking user turns. It also implements robust argument escaping and path resolution to handle Windows-specific command-line parsing and npm shim layouts securely.
plugin/scripts · high confidence
ruflo-market-data plugin v0.3.4: security guard, namespace routing fix, and mod integration
The ruflo-market-data plugin introduces a tighten-only security guard (defaulting to on) that refuses writes to market namespaces containing secrets or credentialed feed URLs, ensuring credentials never enter market memory. It fixes a functional bug where candle history was incorrectly routed to the small \ruvllm\hnsw\\\ prototype router, switching to the \market-data\ namespace for storage and retrieval. The plugin is now registered as a mod (ADR-445), providing a \/market-mod\ command for status, tool connectivity, and secret scanning, alongside a status file for console integration. Documentation and skills have been updated to reflect the correct \memory\\*\ namespace-routed tools and the \embeddings\_generate\ tool name.
plugins/ruflo-market-data · high confidence
ruflo-migrations plugin v0.3.4: namespace-routing fix and tighten-only security guard
The ruflo-migrations plugin (v0.3.4) fixes a namespace-routing bug where migration metadata was incorrectly stored using tier-routed tools (\agentdb\hierarchical-\\) instead of namespace-routed tools (\memory\\\), ensuring migration history is correctly tracked. It also introduces a tighten-only security guard (default on) that refuses memory writes containing secrets or database URLs with inline passwords, and adds a local \/migrations-mod\ command for scanning SQL for destructive operations and secrets without model turns.
plugins/ruflo-migrations, plugins/ruflo-observability · high confidence
Test coverage
1 commit adding/updating tests in plugins/ruflo-agntcy/tests; 10 commits adding/updating tests in plugins/ruflo-adr/scripts/\_\tests\\_; 14 commits adding/updating tests in plugins/ruflo-agentdb/tests; 2 commits adding/updating tests in plugins/ruflo-adr/tests; Add smoke test script for plugin contract validation; Added V3 test fixtures and helper utilities; Added integration tests for BBS and ChatGPT federation plugin guards; Added smoke test script for plugin contract validation; Added structural smoke test for ruflo-intelligence plugin; Added structural smoke test for ruflo-intelligence plugin; Added test scripts for the IPFS-based plugin store; Added tests for OAuth 2.1 federation authentication and publisher security; Added tests for autopilot mod configuration and security guard; Added tests for browser security guard and status command; Added tests for federation mod secret and PII guards; Added tests for plugin registry, collection management, and SQL injection hardening; Added tests for the DAA mod secret guard and status file; Added tests for the DDD mod secret guard and status commands; Added tests for the aidefence-mod plugin; Added tests for the ruflo-docs secret guard and /docs-mod command; Expanded test coverage for core subsystems and CI workflows; New integration test suite for agent lifecycle, SONA mode resolution, and token optimization metrics; New performance benchmarks for console subsystems; Post-deployment e2e and structural smoke tests for ChatGPT Federation; Smoke test script validates plugin contract and federation mod compliance; Structural smoke test for ruflo-autopilot v0.3.2; Test coverage for ADR management and confirm-gated actions.
Dependencies
Introduces root Cargo workspace and adds ruflo-arena plugin
The repository now includes a root \Cargo.toml\ that defines a Rust workspace containing the \ruflo-federation-peer\ and \ruflo-agntcy\ crates, enabling Rust tooling to resolve these components. Additionally, a new \ruflo-arena\ plugin package is added, providing a CLI and library for competitive swarm evaluation (arenas and tournaments).
(dependencies) · high confidence
Written by cca.canine.dev from the codebase's own history, inside the signed delivery this page is composed from.
How this codebase got here
Score
- CAI 49 → 51 (+2.5)
- Rubric changed (rubric-2026.10.5 → rubric-2026.11.4) — scores are not directly comparable.
Lenses
- Code Health 52 → 50 (-2.2)
- Architecture 49 → 49 (+0.2)
- Maturity 88 → 89 (+1.8)
- Readiness 45 → 52 (+7.4)
- Security 50 → 52 (+2.4)
- Domain Modelling 76 → 79 (+2.3)
- Accessibility 51 → 51 (+0.3)
- Performance 73 → 73 (+0.0)
Resolved (813)
- ADR not followed: ADR 460: Mission links, run events, mission-as-workflow and guidance on the Workflows page (v3/docs/adr/ADR-460-workflow-links-nesting-and-guidance.md)
- ADR superseded but its status says otherwise (v3/docs/adr/ADR-074-self-learning-wiring-2245.md)
- Classify ADR enforcement: ADR 398: Planner Safe Agent Registration Descriptors (v3/docs/adr/ADR-398-planner-safe-agent-registration-descriptors.md)
- Classify ADR enforcement: ADR 399: Repository Local Git Execution Boundary (v3/docs/adr/ADR-399-repository-local-git-execution-boundary.md)
- Classify ADR enforcement: ADR 404: Ruflo as a Claude Code Mod (Function Hooks) (v3/docs/adr/ADR-404-claude-code-mods-function-hooks.md)
- Classify ADR enforcement: ADR 405: ruOS Desktops as Swarm Execution Hosts (v3/docs/adr/ADR-405-ruos-desktops-as-swarm-hosts.md)
- Classify ADR enforcement: ADR 406: Ruflo mission control through compatible Claude Code mods (v3/docs/adr/ADR-406-ruflo-mod-command-compatibility.md)
- Classify ADR enforcement: ADR 407: The ruflo-console cockpit: interface, AI terminal and safety contract (v3/docs/adr/ADR-407-ruflo-console-cockpit.md)
- Classify ADR enforcement: ADR 408: Mission Control in the ruflo-console cockpit (v3/docs/adr/ADR-408-ruflo-console-mission-control.md)
- Classify ADR enforcement: ADR 409: The Mission Control planner and the one lifecycle (v3/docs/adr/ADR-409-console-mission-planner-and-lifecycle.md)
- Classify ADR enforcement: ADR 410: Claude's guidance on a mission, and the AIDefence screen (v3/docs/adr/ADR-410-console-claude-guidance-and-aidefence.md)
- Classify ADR enforcement: ADR 411: The Claude UI bridge: Ask Claude, Launch, and the plugin map (v3/docs/adr/ADR-411-console-claude-ui-bridge-ask-launch-plugin-map.md)
- Classify ADR enforcement: ADR 412: Answers open where they were asked (v3/docs/adr/ADR-412-console-attention-answers-open-where-asked.md)
- Classify ADR enforcement: ADR 413: The ruflo-console regression system (v3/docs/adr/ADR-413-console-regression-system.md)
- Classify ADR enforcement: ADR 414: The Loop Manager (v3/docs/adr/ADR-414-console-loop-manager.md)
- Classify ADR enforcement: ADR 415: The Learning page and the ruflo Optimizer (v3/docs/adr/ADR-415-console-learning-page-and-optimizer.md)
- Classify ADR enforcement: ADR 416: The Timeline and Events pages (v3/docs/adr/ADR-416-console-timeline-and-events.md)
- Classify ADR enforcement: ADR 417: The AgentBBS section of the x.ruv.io view (v3/docs/adr/ADR-417-console-agentbbs-section.md)
- Classify ADR enforcement: ADR 418: Tool rows name the mission task they belong to (v3/docs/adr/ADR-418-console-tool-row-task-ownership.md)
- Classify ADR enforcement: ADR 419: Three more AgentDB reads in the Memory Lab, and a coverage audit of the knowledge and automation surfaces (v3/docs/adr/ADR-419-console-agentdb-reads-and-coverage-audit.md)
- …and 793 more
New (1064)
- ADR lacks an enforcement field (v3/docs/adr/ADR-502-verified-outcome-learning-loop.md)
- ADR lacks an enforcement field (v3/docs/adr/ADR-503-learning-loop-gaming-defences.md)
- ADR lacks an enforcement field (v3/docs/adr/ADR-504-mods-real-binary-test-tier.md)
- ADR lacks an enforcement field (v3/docs/adr/ADR-505-compaction-snapshot-and-reload.md)
- ADR lacks an enforcement field (v3/docs/adr/ADR-506-plugin-userconfig-secrets.md)
- ADR lacks an enforcement field (v3/docs/adr/ADR-507-plugin-footprint-skills-commands.md)
- ADR lacks an enforcement field (v3/docs/adr/ADR-508-cost-ladder-gateway-budgets.md)
- ADR lacks an enforcement field (v3/docs/adr/ADR-509-plugin-trust-signing-and-scan.md)
- ADR lacks an enforcement field (v3/docs/adr/ADR-510-headless-and-sdk-plugin-staging.md)
- ADR lacks an enforcement field (v3/docs/adr/ADR-511-learning-loop-actionable-lessons.md)
- ADR lacks an enforcement field (v3/docs/adr/ADR-512-learning-loop-on-classic-hooks.md)
- ADR lacks an enforcement field (v3/docs/adr/ADR-513-ecosystem-stable-versions-and-dependency-convergence.md)
- ADR lacks an enforcement field (v3/docs/adr/ADR-514-agentdb-3-0-0-stable.md)
- ADR lacks an enforcement field (v3/docs/adr/ADR-515-agentic-flow-3-0-0-line-reconciliation.md)
- ADR lacks an enforcement field (v3/docs/adr/ADR-516-aidefence-source-recovery.md)
- ADR lacks an enforcement field (v3/docs/adr/ADR-517-aidefence-aimds-pattern-packs.md)
- ADR lacks an enforcement field (v3/docs/adr/ADR-518-external-agentdb-agentic-flow-plugins.md)
- ADR lacks an enforcement field (v3/docs/adr/ADR-519-agentdb-3-2-episode-recall.md)
- ADR lacks an enforcement field (v3/docs/adr/ADR-520-agentdb-3-3-episode-crud.md)
- ADR lacks an enforcement field (v3/docs/adr/ADR-521-bounded-context-delegation.md)
- …and 1044 more
Changes since last survey
- 300 commits — 175 feature/other, 125 fixes
By area
- v3/@claude-flow — 120 commits
- (repo) — 109 commits
- (root) — 21 commits
- v3/plugins — 7 commits
- plugins/ruflo-console — 5 commits
- plugins/ruflo-metaharness — 5 commits
- docs/dream-cycle — 4 commits
- plugins/ruflo-core — 4 commits
- plugins/ruflo-graph-intelligence — 3 commits
- plugins/ruflo-mods — 3 commits
- ruflo/src — 3 commits
- .github/workflows — 2 commits
- plugins/ruflo-agentdb — 2 commits
- v3/docs — 2 commits
- plugins/ruflo-browser — 1 commit
- plugins/ruflo-cost-tracker — 1 commit
- plugins/ruflo-federation — 1 commit
- plugins/ruflo-intelligence — 1 commit
- plugins/ruflo-plugin-creator — 1 commit
- plugins/ruflo-ruvector — 1 commit
Notable commits
- fix: Merge branch 'main' into fix/4744-helper-integrity
- fix: Merge branch 'main' into fix/daemon-queue-supervisor-gate
- fix: Merge pull request #4519 from ruvnet/fix/daemon-queue-supervisor-gate
- fix: Merge pull request #4821 from lj1166/fix/4754-bound-imports-and-join
- fix: Merge pull request #4827 from ruvnet/fix/4711-cost-health-cold-npx
- fix: Merge pull request #4829 from MannXo/fix/4748-federation-require-jcs
- fix: Merge pull request #4830 from MentaNew/fix/embeddings-init-zero
- fix: Merge pull request #4847 from lj1166/fix/4744-helper-integrity
- fix: Merge pull request #4850 from Tong-bit-art/fix/hive-mind-permissions-default
- fix: Merge pull request #4853 from NotZhu/fix/4734-readonly-getentry
- fix: Merge pull request #4859 from PerryLink/fix/cleanup-preserve-user-data-4833-20261106
- fix: Merge pull request #4877 from PerryLink/fix/providers-test-openai-endpoint-4834-20261107
- fix: Merge pull request #4879 from NotZhu/fix/transfer-store-download-hardening
- fix: Merge pull request #4900 from ruvnet/fix/3.74.1-stale-helpers
- fix: Merge pull request #4907 from lj1166/fix/4755-safe-defaults
- fix: Merge pull request #4908 from ruvnet/fix/4902-queen-failover-config
- fix: Merge pull request #4909 from ruvnet/fix/4888-atomic-short-write
- fix: Merge pull request #4912 from ruvnet/fix/tool-schema-limits
- fix: Merge pull request #4913 from ruvnet/fix/4880-4876-persist-and-search
- fix: Merge pull request #4914 from ruvnet/fix/4862-aidefence-learn-pii
- …and 280 more
Written by cca.canine.dev from the codebase's own history, inside the signed delivery this page is composed from.
Survey your own repository
ruvnet/ruflo was measured the same way every project in this corpus was: the same rubric, at a pinned commit, with the result published in full. Point a surveyor at a repository you know and see whether you agree with it.
About this page
- The score is its most recent published measurement, taken on 14 November 2026 at a pinned commit. It is not a live figure and does not change until the project is measured again.
- Measured at commit 3a1479883b55b7f786ef0bf3141478e4dfb9444f — the exact code this score is about.
- Scored under rubric-2026.11.4 — the same rubric and the same method as every other entry in this index.
- Measured by Canine Code Analyzer (cca.canine.dev) using cca measure-2026.11.65+preprod-e1a3b69fc8ee.