Hexis vs Ruflo: Features, Pricing & Which Is Better (2026)

A side-by-side comparison of Hexis and Ruflo — features, pricing, and ideal use cases — to help you decide which AI tool fits your workflow.

Hexis

Bevelites GmbH

Freemium

Open-source git-backed control plane for enterprise AI agents — context, skills, tools and permissions as files in your own repo, served over MCP.

Git-backed source of truth: Context, skills, tools, permissions, and agent identities live as Markdown/YAML files in your own repository — reviewed, branched, and diffed like any other code.
Typed knowledge with provenance: Every fact is a typed node with source, owner, and last verification date, compiled into a graph you can traverse, mass-update, and dashboard.
Markdown skills, not prompt fragments: Procedures live as readable Markdown files that domain owners can review, instead of prompt snippets buried in a vendor config.
Tool manifests with vaulted secrets: Declare each tool once, hold secrets in a vault, and set file-level access rules that say which agent may read which file and call which endpoint.
Per-agent identity: Each agent is a named actor with its own credentials and scope — no shared service accounts, and every action is attributable to a specific agent.
Any runtime via MCP or UTCP: Serves the same governed surface to Claude Code, Cursor, ChatGPT, opencode, background agents, in-platform agents, and self-hosted models.
Open-source Apache-friendly release: Hexis is the OSS core; Bevel (paid) adds hosted services, integrations, and enterprise support on top.

Enterprises running AI agents across multiple vendors (Claude, Cursor, ChatGPT) who want context and skills defined once in their own infrastructure instead of re-uploaded per vendor.
Procurement, GTM, and RFI teams building specialised agents (e.g., tender desks, campaign agents) on top of a shared reusable knowledge layer with provenance.
Platform teams that need per-agent identities and file-level access controls so every tool call is attributable and destructive endpoints are gated by review.
Companies wary of vendor lock-in who want a portable spec (MCP/UTCP) so they can switch agent runtimes without rebuilding context, skills, and tool wiring.
Engineering leaders who want the AI 'operating manual' reviewed in git diffs and change requests instead of edited inside a black-box vendor console.
Open-source-first teams evaluating a control plane locally with Hexis before committing to Bevel's hosted platform for larger deployments.

rUv (ruvnet)

Free

Open-source agent meta-harness for Claude Code and Codex adding swarms, self-learning memory, MCP tools and security guardrails.

Agent Swarms: Over 100 specialized agents self-organize into coordinated swarms to split up and run complex tasks.
Self-Learning Memory: Remembers across sessions and learns from successful trajectories using local vector retrieval and contrastive updates.
MCP Server: Run npx ruflo mcp start to expose Ruflo's tools to Codex or any MCP-enabled client.
Claude Code Plugins: Core tools, a visual console and runtime mods add routing and policy enforcement inside Claude Code.
Terminal UI Console: Shows Overview, Cost, Swarm, Events and Missions pages, and also runs standalone.
Federation: Lets agents on different machines communicate securely without leaking data.
Hook-Based Routing: Hooks automatically route tasks and coordinate agents in the background.

Large Refactors: Split a big codebase change across a swarm of specialist agents.
Persistent Project Context: Keep knowledge across Claude Code sessions with vector memory.
Multi-Client Setups: Connect Codex or other MCP clients to the same tool server.
Cost Monitoring: Track spend and agent activity from the console during long runs.
Multi-Machine Collaboration: Let agents on separate machines coordinate through federation.