Ruflo
Marketplace AktivRuFlo Marketplace: Claude Code native agents, swarms, workers, and MCP tools for continuous software engineering
To provide a robust framework for orchestrating complex multi-agent AI workflows, enabling advanced collaboration, self-learning, and secure communication for continuous software engineering tasks within Claude Code.
Funktionen
- Multi-agent collaboration and swarm coordination
- Self-learning memory and knowledge graph capabilities
- Secure agent federation across machines and organizations
- Extensive plugin ecosystem for specialized tasks
- Support for multiple LLM providers and local models
Anwendungsfälle
- Automating complex software engineering workflows with coordinated agents
- Enabling agents to learn from past tasks and improve performance
- Securely integrating agents across different teams or trust boundaries
- Building intelligent systems that can autonomously plan and execute goals
Nicht-Ziele
- Replacing core Claude Code functionality
- Providing a simple command-line tool without orchestration
- Allowing agents to operate without any form of memory or learning
Scope
- info:Single responsibility principleWhile Ruflo aims to provide a comprehensive platform for multi-agent AI orchestration, the sheer number of plugins and their diverse functionalities (from code quality to IoT) could be seen as an aggregation of loosely related workflows, though they are unified under the umbrella of agentic software engineering.
- info:Minimal I/O surfaceThe README describes many capabilities but does not detail the specific input schemas or response shapes for individual tools or agents, making it difficult to assess the minimality of their I/O surfaces.
- warning:Tool surface sizeThe README indicates an extremely large surface area with '100+ Agents', '60+ commands', '30 skills', and '32 plugins', far exceeding the recommended target of 3-10 tools, which could lead to complexity and invocation ambiguity.
- info:Dry-run previewWhile the documentation emphasizes security and meticulous workflows, specific `--dry-run` flags or equivalent preview mechanisms for all state-changing operations are not explicitly detailed.
Invocation
- warning:Scoped toolsThe README mentions over 100 agents and numerous commands, with specific plugins like `ruflo-ruvector` listing 103 tools. While many are likely specific verb-noun actions, the sheer volume and the potential for generalist commands within plugins suggests a risk of less-than-ideal tool scoping.
- warning:Overlapping near-synonym toolsWith over 100 agents and numerous tools, it's highly probable that near-synonym tools exist for similar actions (e.g., various search or analysis commands across plugins), forcing the model to disambiguate frequently.
Documentation
- info:Configuration & parameter referenceThe README provides extensive information on installation and capabilities, but detailed documentation for all parameters and configuration options across all plugins is not explicitly detailed here, with references to external User Guide sections.
Trust
- warning:Issues AttentionThe repository has 68 open issues and 373 closed issues in the last 90 days, indicating a high volume of activity. However, the ratio of open to closed issues suggests that maintainer response time might be a concern, potentially indicating slower resolution of reported bugs or feature requests.
Code Execution
- info:ValidationWhile the documentation highlights security and validation (e.g., PII detection, prompt injection defense), it does not explicitly mention the use of schema validation libraries like Zod or Pydantic for input arguments or structured output.
- info:Error HandlingThe documentation emphasizes robust error handling and security, but does not explicitly detail the structured error reporting (code, message, retryable, hint) for all agent interactions or tool failures.
- info:LoggingThe project mentions audit trails for Federation and structured logs/traces/metrics for observability, but it's not clear if destructive actions or outbound calls are logged to a local audit file by default for all operations.
Errors
- info:Actionable error messagesWhile the documentation stresses security and robustness, specific details on the structure and content of user-facing error messages (what/why/next-step) for all potential failure paths are not explicitly detailed.
Protocol
- info:Idempotent retry & timeoutsThe documentation highlights robust error handling and security, but does not explicitly detail whether mutating operations are idempotent or if hard timeouts are enforced for all internal agent communications and tool calls.
Installation
/plugin marketplace add ruvnet/rufloEnthält 32 Erweiterungen
Plugin (32)
Foundation plugin — registers the ruflo MCP server (300+ tools across memory/agentdb/embeddings/hooks/aidefence/neural/autopilot/browser/agent/swarm), provides 3 generalist agents (coder/researcher/reviewer), 3 first-run skills, and a curated plugin-discovery catalog
Agent teams, swarm coordination, Monitor streams, and worktree isolation — wraps 4 swarm_* + 8 agent_* MCP tools (12 total) plus 6 topologies (hierarchical / mesh / hierarchical-mesh / ring / star / adaptive)
Cache-aware /loop workers and CronCreate background automation — wraps 5 hooks_worker-* MCP tools (list/dispatch/status/detect/cancel) and exposes 12 background worker triggers (ultralearn, optimize, consolidate, predict, audit, map, preload, deepdive, document, refactor, benchmark, testgaps)
Security review, dependency scanning, policy gates, and CVE monitoring
RuVector memory with HNSW search, AgentDB, and semantic retrieval
Test gap detection, coverage analysis, and automated test generation — drives the testgaps background worker via hooks_worker-dispatch; SPARC Refinement-phase canonical owner
Documentation generation, API docs (JSDoc/TSDoc/OpenAPI), and drift detection — drives the `document` background worker via hooks_worker-dispatch; uses Haiku model for cost-efficient docs work
Autonomous /loop-driven task completion with learning, prediction, and progress tracking — wraps 10 autopilot_* MCP tools (status/enable/disable/config/reset/log/progress/learn/history/predict)
User-facing surface for Ruflo's self-learning system: 6 neural_* + 10 hooks_intelligence_* + 9 routing/meta hooks + 4 SONA/MicroLoRA tools (29 total). Implements the 4-step pipeline (RETRIEVE → JUDGE → DISTILL → CONSOLIDATE) and IPFS-based cross-project pattern transfer.
Substrate plugin for Ruflo memory: AgentDB controller bridge (15 agentdb_* MCP tools), RuVector ONNX embeddings (10 embeddings_* tools incl. RaBitQ 32x quantization), and WASM HNSW pattern router (3 ruvllm_hnsw_* tools)
AI safety scanning, PII detection, prompt injection defense, and adaptive threat learning
Session-as-skill browser automation: Playwright + RVF cognitive containers + ruvector trajectories + AgentDB selector memory + AIDefence PII/injection gates
Advanced git workflows with diff analysis, risk scoring, change classification (feature/bugfix/refactor/...), and reviewer recommendations — wraps 6 analyze_* MCP tools (diff, diff-risk, diff-classify, diff-reviewers, file-risk, diff-stats)
Agent runtimes for ruflo — local WASM-sandboxed agents (rvagent: 10 wasm_agent_*/wasm_gallery_* MCP tools, built on @ruvector/rvagent-wasm + @ruvector/ruvllm-wasm per ADR-070) plus Anthropic Claude Managed Agents as a cloud backend (managed_agent_* MCP tools per ADR-115). One interface, local-vs-cloud runtimes.
Workflow automation with templates, orchestration, and lifecycle management — wraps 10 workflow_* MCP tools (create/run/execute/status/list/pause/resume/cancel/delete/template) with full state-machine lifecycle (created → running ↔ paused → completed/cancelled)
Dynamic Agentic Architecture — 8 daa_* MCP tools for adaptive agents (create/adapt), cognitive patterns, workflows (create/execute), knowledge sharing, and learning/performance metrics. Feeds the JUDGE phase of the 4-step intelligence pipeline.
RuVLLM local inference with chat formatting (Claude/GPT/Gemini/Ollama/Cohere), model configuration, MicroLoRA fine-tuning, and SONA real-time adaptation
RVF format for portable agent memory, session persistence, and cross-platform transfer
Scaffold, validate, and publish new Claude Code plugins with the canonical plugin contract — ADR + smoke + Compatibility + namespace coordination + MCP-tool drift warnings
Long-horizon goal planning, deep research orchestration, and adaptive replanning using GOAP algorithms
ADR lifecycle management — create, index, supersede, check compliance, and link Architecture Decision Records to code via AgentDB hierarchical store + causal edges (supersedes/amends/depends-on/related)
Token usage tracking, model cost attribution per agent, budget alerts, and optimization recommendations — uses memory_* (namespace-routed) for cost-tracking and cost-patterns; pairs with federation budget circuit breaker (ADR-097)
Domain-Driven Design scaffolding — bounded contexts, aggregate roots, domain events, value objects, repositories, and anti-corruption layers; navigable domain graph stored in AgentDB
Cross-installation agent federation with zero-trust security, peer discovery, consensus-based task routing, and per-call budget circuit breaker (ADR-097)
IoT device lifecycle, telemetry anomaly detection, fleet management, and witness chain verification for Cognitum Seed hardware
Knowledge graph construction — entity extraction, relation mapping, and pathfinder graph traversal
Market data ingestion — feed normalization, OHLCV vectorization, and HNSW-indexed pattern matching
Schema migration management — generate, validate, dry-run, and rollback database migrations
Neural trading via npx neural-trader — self-learning strategies, Rust/NAPI backtesting, 112+ MCP tools, swarm coordination, and portfolio optimization
Structured logging, distributed tracing, and metrics — correlate agent swarm activity with application telemetry
Self-learning vector database via npx ruvector@0.2.25 — HNSW, adaptive LoRA embeddings, code-graph clustering, hooks routing, brain/SONA, 103 MCP tools
SPARC methodology — Specification, Pseudocode, Architecture, Refinement, Completion phases with gate checks
Qualitätspunktzahl
Vertrauenssignale
Ähnliche Erweiterungen
Claude Recap Marketplace
100Themenbasierter automatischer Speicher für Claude Code
Cortex Plugins
100Persistent memory and cognitive profiling plugins for Claude Code
Claude Code Plugins
99Production-ready workflow orchestration with 80 focused plugins, 185 specialized agents, and 153 skills - optimized for granular installation and minimal token usage
Compound Engineering
99Cursor plugin marketplace for Every Inc plugins
Crowdcast
99Der Crowdcast Multi-Agent-Social-Simulation-Plugin für den TheQmaks Claude Code Marktplatz.
Claude Code Skills Marketplace
75Claude Code Skills for software engineering workflows - Git automation, testing, and code review