Skip to content

armgit / ruflo

Public Imported from GitHub
0

🌊 The original agent harness. Deploy intelligent multi-player swarms, coordinate autonomous workflows, and build conversational AI systems. Features adaptive memory, self-learning intelligence, federation, vector RAG integration, and native Claude Code / Codex / Hermes and many more Integrated

docs(readme): move Web UI section after What You Get + emoji feature table

Reuvene89a3a20d912
diff --git a/README.md b/README.md
index c610748a54..5c79e544df 100644
--- a/README.md
+++ b/README.md
@@ -157,54 +157,58 @@ npm install -g ruflo@latest
claude mcp add ruflo -- npx -y @claude-flow/cli@latest
```
+---
+
+## What You Get
+
+| Capability | Description |
+|------------|-------------|
+| πŸ€– **100+ Agents** | Specialized agents for coding, testing, security, docs, architecture |
+| πŸ“‘ **Comms Layer** | Zero-trust federation β€” agents across machines/orgs discover, authenticate, and exchange work securely |
+| 🐝 **Swarm Coordination** | Hierarchical, mesh, and adaptive topologies with consensus |
+| 🧠 **Self-Learning** | SONA neural patterns, ReasoningBank, trajectory learning |
+| πŸ’Ύ **Vector Memory** | HNSW-indexed AgentDB with 150x-12,500x faster search |
+| ⚑ **Background Workers** | 12 auto-triggered workers (audit, optimize, testgaps, etc.) |
+| 🧩 **Plugin Marketplace** | 32 native Claude Code plugins + 21 npm plugins |
+| πŸ”Œ **Multi-Provider** | Claude, GPT, Gemini, Cohere, Ollama with smart routing |
+| πŸ›‘οΈ **Security** | AIDefence, input validation, CVE remediation, path traversal prevention |
+| 🌐 **Agent Federation** | Cross-installation agent collaboration with zero-trust security |
+| πŸ’¬ **[Web UI Beta](https://flo.ruv.io/)** | Multi-model chat at flo.ruv.io with parallel MCP tool calling and an in-browser WASM tool gallery |
+
<p align="center">
<a href="https://flo.ruv.io/">
- <img src="v3/docs/assets/ruVocal-welcome.png" alt="RuFlo Web UI welcome screen at flo.ruv.io β€” multi-agent AI chat with MCP tools, Model Context Protocol orchestration, persistent vector memory (AgentDB + HNSW), parallel tool calling, and 6 frontier models including Qwen 3.6 Max, Claude Sonnet 4.6, Gemini 2.5 Pro, and GPT-4o" width="520" />
+ <img src="v3/docs/assets/ruVocal.png" alt="RuFlo Web UI executing parallel MCP tool calls at flo.ruv.io β€” ruflo__memory_store and ruflo__memory_search firing in a single model turn with the 'Step 1 β€” 2 tools completed' parallel-execution indicator, thinking process panel visible, Qwen 3.6 Max as the active model. Multi-agent AI chat with Model Context Protocol (MCP) tool calling, persistent vector memory via AgentDB + HNSW, swarm coordination, and 6 frontier models including Claude Sonnet 4.6, Gemini 2.5 Pro, and GPT-4o through OpenRouter." width="100%" />
</a>
</p>
-### Web UI (Beta) β€” [flo.ruv.io](https://flo.ruv.io/)
+### Web UI (Beta) β€” [flo.ruv.io](https://flo.ruv.io/) Β· [ruflo.ruv.io](https://ruflo.ruv.io/) Β· [ruvocal.ruv.io](https://ruvocal.ruv.io/)
-**RuFlo's web UI is a multi-model AI chat with built-in Model Context Protocol (MCP) tool calling.** Talk to Qwen, Claude, Gemini, or GPT-4o while RuFlo invokes the same ~210 MCP tools the CLI uses β€” agent orchestration, persistent vector memory, swarm coordination, code analysis, GitHub ops β€” directly from the chat. No install, no API key required to try it.
+**RuFlo's web UI is a multi-model AI chat with built-in Model Context Protocol (MCP) tool calling.** Talk to Qwen, Claude, Gemini, or GPT-4o while RuFlo invokes the same MCP tools the CLI uses β€” agent orchestration, persistent memory, swarm coordination, code review, GitHub ops β€” directly from chat. No install, no API key needed to try it.
-- **6 frontier models**: Qwen 3.6 Max (default), Claude Sonnet 4.6, Claude Haiku 4.5, Gemini 2.5 Pro, Gemini 2.5 Flash, GPT-4o β€” all via OpenRouter
-- **~210 MCP tools** across 5 server groups (Core, Intelligence, Agents, Memory, DevTools) plus an 18-tool in-browser WASM gallery
-- **Parallel tool calling**: a single model response can fan out 4–6+ `tool_calls` simultaneously. The UI streams each tool's result as a card with a *Step N β€” M tools completed* parallel indicator
-- **Persistent memory** across sessions via AgentDB + HNSW vector search
-- **Help icon β†’ "RuFlo Capabilities"** modal with the full tool catalog, model lineup, and quick-start
+| | What it is | Why it matters |
+|---|------------|----------------|
+| 🧠 | **6 frontier models** | Qwen 3.6 Max (default), Claude Sonnet 4.6, Claude Haiku 4.5, Gemini 2.5 Pro, Gemini 2.5 Flash, GPT-4o β€” switch any time. One key (OpenRouter) covers them all. |
+| πŸ› οΈ | **~210 tools, ready to call** | 5 server groups (Core, Intelligence, Agents, Memory, DevTools) plus an 18-tool gallery that runs entirely in your browser β€” works offline. |
+| ⚑ | **Tools run in parallel** | One model response can fire 4–6+ tools at the same time. The UI shows them as cards with a *Step 1 β€” 2 tools completed* badge so you can see exactly what ran. |
+| πŸ’Ύ | **Memory that sticks** | Say *"remember my favorite color is indigo"* and ask weeks later β€” RuFlo recalls it. Backed by AgentDB + HNSW vector search (β‰₯150Γ— faster than brute force). |
+| πŸ“˜ | **Built-in capabilities tour** | Click the question-mark icon in the sidebar β€” a "RuFlo Capabilities" modal opens with the full tool list, model strengths, architecture, and keyboard shortcuts. |
+| 🌍 | **Three URLs, one app** | [`flo.ruv.io`](https://flo.ruv.io/), [`ruflo.ruv.io`](https://ruflo.ruv.io/), [`ruvocal.ruv.io`](https://ruvocal.ruv.io/) β€” all aliased to the same Cloud Run service through Cloudflare. |
+| πŸš€ | **Zero install** | Open the URL, pick a model, type a question. That's the whole onboarding. |
<details>
-<summary><strong>See it in action</strong> β€” parallel <code>memory_store</code> + <code>memory_search</code> with thinking process visible</summary>
+<summary><strong>Welcome screen preview</strong> β€” RuFlo branding and entry modal</summary>
<p align="center">
<a href="https://flo.ruv.io/">
- <img src="v3/docs/assets/ruVocal.png" alt="RuFlo Web UI executing parallel MCP tool calls β€” ruflo__memory_store and ruflo__memory_search firing in one model turn with the 'Step 1 β€” 2 tools completed' parallel-execution indicator, thinking process panel visible, Qwen 3.6 Max active model" width="100%" />
+ <img src="v3/docs/assets/ruVocal-welcome.png" alt="RuFlo Web UI welcome modal β€” RuFlo Intelligent Workflows branding with MCP Tools badge, welcome copy, and Start chatting CTA" width="520" />
</a>
</p>
</details>
-Open it directly: **[https://flo.ruv.io/](https://flo.ruv.io/)** (also `ruflo.ruv.io` / `ruvocal.ruv.io`). See [ADR-033](ruflo/docs/adr/ADR-033-RUVOCAL-WASM-MCP-INTEGRATION.md) for the architecture and [issue #1689](https://github.com/ruvnet/ruflo/issues/1689) for the roadmap.
+Open it directly: **[https://flo.ruv.io/](https://flo.ruv.io/)**. See [ADR-033](ruflo/docs/adr/ADR-033-RUVOCAL-WASM-MCP-INTEGRATION.md) for the architecture and [issue #1689](https://github.com/ruvnet/ruflo/issues/1689) for the roadmap.
-> **Keywords**: AI chat UI Β· MCP server Β· Model Context Protocol Β· multi-agent orchestration Β· Claude Β· Gemini Β· Qwen Β· GPT-4o Β· vector memory Β· HNSW Β· agent swarms Β· parallel tool calling Β· LLM tool use Β· OpenRouter
-
----
-
-## What You Get
-
-| Capability | Description |
-|------------|-------------|
-| πŸ€– **100+ Agents** | Specialized agents for coding, testing, security, docs, architecture |
-| πŸ“‘ **Comms Layer** | Zero-trust federation β€” agents across machines/orgs discover, authenticate, and exchange work securely |
-| 🐝 **Swarm Coordination** | Hierarchical, mesh, and adaptive topologies with consensus |
-| 🧠 **Self-Learning** | SONA neural patterns, ReasoningBank, trajectory learning |
-| πŸ’Ύ **Vector Memory** | HNSW-indexed AgentDB with 150x-12,500x faster search |
-| ⚑ **Background Workers** | 12 auto-triggered workers (audit, optimize, testgaps, etc.) |
-| 🧩 **Plugin Marketplace** | 32 native Claude Code plugins + 21 npm plugins |
-| πŸ”Œ **Multi-Provider** | Claude, GPT, Gemini, Cohere, Ollama with smart routing |
-| πŸ›‘οΈ **Security** | AIDefence, input validation, CVE remediation, path traversal prevention |
-| 🌐 **Agent Federation** | Cross-installation agent collaboration with zero-trust security |
-| πŸ’¬ **[Web UI Beta](https://flo.ruv.io/)** | Multi-model chat at flo.ruv.io with parallel MCP tool calling and an in-browser WASM tool gallery |
+> **Keywords**: AI chat UI Β· MCP server Β· Model Context Protocol Β· multi-agent orchestration Β· Claude Β· Gemini Β· Qwen Β· GPT-4o Β· vector memory Β· HNSW Β· agent swarms Β· parallel tool calling Β· LLM tool use Β· OpenRouter Β· AgentDB Β· semantic search Β· ruvocal Β· ruflo.ruv.io
### Agent Federation β€” Slack for Agents
diff --git a/ruflo/src/ruvocal/src/lib/components/RufloHelpModal.svelte b/ruflo/src/ruvocal/src/lib/components/RufloHelpModal.svelte
index 29719b74f6..9614ec0cc9 100644
--- a/ruflo/src/ruvocal/src/lib/components/RufloHelpModal.svelte
+++ b/ruflo/src/ruvocal/src/lib/components/RufloHelpModal.svelte
@@ -153,6 +153,41 @@
'"Run ruflo__system_status, ruflo__performance_metrics, and ruflo__memory_list in parallel and summarize."',
'"Spawn a 5-agent hierarchical swarm (architect, coder, tester, reviewer, security-auditor) for a Python→TypeScript refactor."',
'"Use ruvector__hooks_route on the task: add OAuth to a SvelteKit API. Then spawn the recommended agent."',
+ '"Search RuFlo memory for prior decisions about authentication, then web_research recent OAuth2 best practices, in parallel."',
+ '"Analyze the diff at github.com/ruvnet/ruflo/pull/1687 β€” risk score, classify, and suggest reviewers."',
+ ];
+
+ const stack: { name: string; role: string }[] = [
+ { name: "SvelteKit + adapter-node", role: "UI runtime, SSR + streaming MCP responses" },
+ { name: "OpenRouter", role: "Single OAI-compatible endpoint, 6 models, smart fail-over" },
+ { name: "ruflo backend (npx ruflo mcp start)", role: "Agents, memory, swarm, devtools (~158 tools)" },
+ { name: "ruvector backend (npx ruvector mcp start)", role: "Intelligence, routing, trajectories (~49 tools)" },
+ { name: "AgentDB (sql.js + HNSW)", role: "Persistent memory, 384-dim ONNX embeddings, 150x–12 500x search speedup" },
+ { name: "rvagent-wasm (in-browser, ~588 KB)", role: "18-tool gallery with IndexedDB persistence, optional Web Worker" },
+ { name: "MongoDB (embedded)", role: "Conversation + session storage (ephemeral on Cloud Run cold starts today)" },
+ { name: "Cloud Run (us-central1)", role: "Both ruvocal chat-ui and mcp-bridge services with min-instance=1 warm" },
+ ];
+
+ const domains: { url: string; note: string }[] = [
+ { url: "flo.ruv.io", note: "Primary β€” shortest URL, recommended" },
+ { url: "ruflo.ruv.io", note: "Brand alias" },
+ { url: "ruvocal.ruv.io", note: "Original alias (matches the upstream chat-ui fork name)" },
+ { url: "ruvocal-875130704813.us-central1.run.app", note: "Raw Cloud Run URL β€” always available, no DNS dependency" },
+ ];
+
+ const shortcuts: { keys: string; action: string }[] = [
+ { keys: "Enter", action: "Send the current message" },
+ { keys: "Shift + Enter", action: "Insert a newline" },
+ { keys: "Esc", action: "Close any open modal" },
+ { keys: "?worker=1 (URL flag)", action: "Run WASM MCP off the main thread (experimental)" },
+ ];
+
+ const cardLegend: { mark: string; meaning: string }[] = [
+ { mark: "βœ“ Called tool", meaning: "Tool returned successfully β€” click the row to expand the JSON result" },
+ { mark: "Step N β€” M tools completed", meaning: "Parallel batch indicator. M tools fired in one model turn via Promise.all" },
+ { mark: "Retry", meaning: "Last response had no content. Click to re-run with the same prompt and model" },
+ { mark: "AUTO (green)", meaning: "Autopilot ON β€” model auto-continues after each tool result" },
+ { mark: "AUTO (gray)", meaning: "Autopilot OFF β€” model stops after each tool result so you can inspect" },
];
let openGroup = $state<string | null>("memory");
@@ -269,6 +304,60 @@
</ul>
</section>
+ <section class="mb-6">
+ <h3 class="mb-2 text-sm font-semibold uppercase tracking-wide text-gray-500">Tool-call cards β€” what the icons mean</h3>
+ <ul class="space-y-1 text-xs">
+ {#each cardLegend as legend}
+ <li class="flex gap-2">
+ <code class="shrink-0 rounded bg-gray-100 px-1.5 py-0.5 font-mono dark:bg-gray-800">{legend.mark}</code>
+ <span class="text-gray-700 dark:text-gray-300">{legend.meaning}</span>
+ </li>
+ {/each}
+ </ul>
+ </section>
+
+ <section class="mb-6">
+ <h3 class="mb-2 text-sm font-semibold uppercase tracking-wide text-gray-500">Under the hood</h3>
+ <p class="mb-2 text-xs text-gray-600 dark:text-gray-400">RuFlo's web UI is a fork of the open-source HuggingFace chat-ui (SvelteKit) wired to a custom MCP bridge that fans out to two backend kernels. Everything runs on Google Cloud Run.</p>
+ <div class="overflow-hidden rounded-lg border border-gray-200 dark:border-gray-700">
+ <table class="w-full text-xs">
+ <tbody class="divide-y divide-gray-200 dark:divide-gray-700">
+ {#each stack as layer}
+ <tr>
+ <td class="whitespace-nowrap px-3 py-1.5 align-top font-mono font-medium">{layer.name}</td>
+ <td class="px-3 py-1.5 align-top text-gray-600 dark:text-gray-400">{layer.role}</td>
+ </tr>
+ {/each}
+ </tbody>
+ </table>
+ </div>
+ </section>
+
+ <section class="mb-6">
+ <h3 class="mb-2 text-sm font-semibold uppercase tracking-wide text-gray-500">Domain aliases β€” pick one, they all work</h3>
+ <ul class="space-y-1 text-xs">
+ {#each domains as d}
+ <li class="flex flex-wrap gap-2">
+ <code class="shrink-0 rounded bg-gray-100 px-1.5 py-0.5 font-mono dark:bg-gray-800">{d.url}</code>
+ <span class="text-gray-600 dark:text-gray-400">{d.note}</span>
+ </li>
+ {/each}
+ </ul>
+ <p class="mt-2 text-[11px] text-gray-500 dark:text-gray-400">Custom domains DNS through Cloudflare unproxied so Google issues + auto-renews the TLS certificate. All four endpoints serve the same Cloud Run revision.</p>
+ </section>
+
+ <section class="mb-6">
+ <h3 class="mb-2 text-sm font-semibold uppercase tracking-wide text-gray-500">Keyboard shortcuts</h3>
+ <ul class="space-y-1 text-xs">
+ {#each shortcuts as s}
+ <li class="flex gap-2">
+ <code class="shrink-0 rounded bg-gray-100 px-1.5 py-0.5 font-mono dark:bg-gray-800">{s.keys}</code>
+ <span class="text-gray-700 dark:text-gray-300">{s.action}</span>
+ </li>
+ {/each}
+ </ul>
+ </section>
+
<section class="mb-6">
<h3 class="mb-2 text-sm font-semibold uppercase tracking-wide text-gray-500">Tips</h3>
<ul class="list-disc space-y-1.5 pl-5 text-sm">