KinetAios

English | 简体中文

🌐 Website → https://phinn.github.io/KinetAios/

Real session: parallel conversations across projects, engine badge per session (Direct / Claude Code / Codex), file & Git & rules tabs, token-cost accounting per turn.

A local-first AI agent dashboard for Windows 11 and macOS — both first-class platforms (the "Win" folder name is historical; the repo is phinn/KinetAios). Platform differences (cmd.exe vs /bin/sh, hotkey, tray) are routed at runtime; features are aligned on both sides. Run multiple agent sessions concurrently with streaming answers, shell/file/search/MCP tools, SQLite history with long-term memory, a global hotkey, and a per-session model. No account, no relay server — your own LLM API key is the only credential.

The 30-second version:

• 🔀 Four engines, one window — Direct (in-house ReAct) + Claude Code + Codex + DeepSeek Harness; switch per session, hand off context via Pipeline
• 🧠 Long-term memory on local SQLite — auto-extracted facts shared across all engines, survives restarts
• 🔒 Local-first, no account — your API key is the only credential; nothing goes through a relay
• 🛠 30+ built-in tools + MCP client/server + plugins — shell, files, web, screenshots, computer use out of the box
• ⌨️ Global-hotkey quick panel — Spotlight for AI agents, summon from anywhere
• 📊 Benchmarked on a real 400k-row business task — in-house engine 9.2 vs Claude Code 7.0 (full report)
• 🖥 Windows 11 & macOS — both first-class

⭐ If KinetAios saves you time, a Star is the loudest thank-you for a solo open-source project.

---

Why KinetAios?

Most AI clients lock you into one provider, drop context when you switch engines, and route your conversations through a relay. KinetAios runs four engines in one window, keeps long-term memory across all of them, and never asks for an account.

| | KinetAios | Claude Desktop | Cherry Studio | Cursor | Codex Desktop | |---|---|---|---|---|---| | Four engines (Direct V1/V2/V3 + Claude Code + Codex + DeepSeek Harness) | ✅ | — | — | — | — | | Local SQLite + automatic long-term memory | ✅ | — | — | — | — | | Cross-engine memory (one user profile shared by all engines) | ✅ | — | — | — | — | | Multiple parallel sessions | ✅ | — | ✅ | — | ✅ | | Global hotkey + quick panel | ✅ | — | — | — | — | | Auto-scan MCP / Skills / Agents | ✅ | ✅ | — | — | — | | Project rules (AGENTS / CLAUDE / KINET) | ✅ | — | — | ✅ | ✅ | | Built-in MCP Server (remote agent control) | ✅ | — | — | — | — | | Plugin system (tools / panels / slash commands) | ✅ | — | — | — | — | | Multimodal (image input + voice + screenshots) | ✅ | ✅ | — | — | — | | Session branching + cross-engine Pipeline orchestration | ✅ | — | — | — | — | | Local-first, no account | ✅ | ✅ | ✅ | — | — |

Install

Grab the latest release:

• Windows — Download KinetAios-Setup-3.9.4.exe (NSIS installer, install → launch → paste your API key)
• macOS — Download the latest .dmg (drag to Applications, same three steps)

Unsigned builds → Windows SmartScreen / macOS Gatekeeper will warn; allow it manually.

First launch: click the ⚙ in the top right → paste your API Key (+ base URL / model, defaults to GLM/Zhipu) → hit Test Connection, then send your first task.

🐋 One key for every model? Pick the OrcaRouter preset — an OpenAI-compatible gateway with 200+ models, adaptive routing, zero markup. Sign up with this link to support the project (5% referral credit).

Built-in OrcaRouter preset:

provider: orcarouter          # Settings → Presets → OrcaRouter (multi-model routing)
protocol: openai              # OpenAI-compatible
base_url: https://api.orcarouter.ai/v1
api_key: sk-orca-...          # create in the OrcaRouter console
model: orcarouter/auto        # adaptive routing, or pin a vendor/model e.g. anthropic/claude-opus-4.8

From source

Requires Node.js 18+ and a network connection (the better-sqlite3 native module needs compiling).

git clone https://github.com/phinn/KinetAios.git
cd KinetAios
npm install      # includes postinstall: rebuild better-sqlite3 for Electron
npm run build
npm start

---

Highlights

Four engines, switchable per session

• Direct V1 (Kaios) — built-in ReAct loop with a dual-protocol provider (OpenAI-compatible & Anthropic, both directions of SSE streaming), tool-level concurrency, sub-agents, context compaction and retries.
• Direct V2 — next-gen ReAct on a Plan-Execute-Verify-Judge architecture, streaming tool calls, step-by-step task lists rendered live in the chat as checklist cards, plus a unified execution loop shared by plan and replan (retry / verification / checkpointing / compaction behave identically on both paths).
• Direct V3 — latest: an intent router picks fast / standard / deep per query; the deep path builds tool calls into a dependency DAG and executes it in parallel — real speedups on multi-step tasks.
• Claude Code — spawns claude -p --output-format stream-json, parses NDJSON, resumes with --resume.
• Codex — spawns codex exec --json, parses JSONL, resumes.
• DeepSeek Harness (3.0+) — spawns the dsh CLI over OpenAI-compatible SSE with built-in OpenAI / Pi-AI providers, retries and token billing. Switchable per session like every other engine.

30+ built-in tools

shell (confirm before running; focus guard — if a command steals the foreground it's handed back automatically), read_file, write_file, edit_file (exact replace), grep, glob, web_fetch (SSRF guard + Jina Reader fallback), web_search (Bing → DuckDuckGo fallback), recall_memory, git_diff (read-only, no confirm), remember_fact / recall_fact (session anchors), memory_replace / memory_append (core memory blocks), dispatch_agent (read-only sub-agent with its own context), spawn_team / team_broadcast / team_send / team_close (multi-agent teams), video_gen (MiniMax H3 text-to-video), todo_write (shared task list, rendered live as checklist cards), and more.

Computer Use (native, zero external dependencies)

• screenshot (with hide_self — the agent's own window turns transparent during the shot, zero focus stealing), screenshot_window (capture any window by title — even occluded/background ones, no desktop switching).
• mouse_click / mouse_scroll / mouse_drag, keyboard_type / keyboard_key — coordinates are converted from the last screenshot's pixel space to the platform's action space (DPI / window-origin aware, with staleness expiry).
• Backed by Electron desktopCapturer + PowerShell (Windows) / cliclick (macOS) / xdotool (Linux). On macOS, bare open is mechanically rewritten to open -gj — the agent never steals your foreground.

MCP integration (client + server)

• Client: the Direct engine auto-connects to MCP servers configured on your system (scans ~/.claude.json, ~/.codex/config.toml, Claude Desktop). stdio transport, auto-reconnect after unexpected drops. The 🔌 button shows connected servers/tools.
• Server: a built-in MCP Server (HTTP+SSE) exposes a run_agent tool — remote machines can invoke your local full agent. Token auth (constant-time compare), 5-minute timeout, zombie-connection detection.

Long-term memory + memory graph

• Each turn, durable "facts about the user" are extracted in the background → SQLite → injected into the next turn. Shared across engines and sessions.
• Memory graph visualization: force-directed graph of memory provenance, conflict detection, timeline. Standalone fullscreen window.
• Import/export memories as JSON (backup or migration).

Skills / Commands / Agents / Plugins

• Scans Claude Code's skills + commands + agents and Codex's skills. Invoke via the / menu or the ⚡ button.
• Skills panel (Settings → Skills): aggregates skills from all third-party sources plus plugin contributions, with search, viewing, and direct source editing (plugin skills read-only). Edits take effect immediately.
• Plugin SDK v3: plugins contribute tools, slash commands, hooks, and fullscreen panels. Injected on demand (keyword matching saves ~60% tokens). 21 built-in plugins: office-suite, brainstorm (Excalidraw), math-practice, cpp-learning, low-altitude (drones), an embedded & IoT suite (arduino-dev / platformio-dev / serial-comm / modbus-dev / mqtt-dev / ble-dev / ota-dev / sensor-lookup / logic-analyzer / hw-diag), nestjs-dev, deepseek-harness, claude-code, codex, marketing-kit (marketing research: competitors / SEO / funnel / app-review mining + a marketing console panel), and more.

Sidebar (left to right)

• ＋ new session.
• 📂 Workbench — project cards grouped by cwd, each showing recent activity + cost. A "Context" button edits KINET-CONTEXT.md.
• Live session status — running sessions show the engine's current step (tool executing / retrying / compacting) right under the title in the sidebar, no need to open the channel.
• 📊 Dashboard — standalone window with live token usage, cost stats, engine distribution.
• 🌐 Files — file browsing + <webview> preview (HTML/SVG/PNG/JPG/PDF) + editor. Multi-tab. The address bar accepts file:// / http(s):// / localhost:<port>.
• 🏘️ Town — game-style isometric visualization of remote nodes (other KinetAios instances) on your network.
• 🧠 Memory — memory panel: current channel / all, inline edit/delete, provenance.
• 🔌 Plugins — plugin management: enable/disable, search & filter, category cards.
• ⚙️ Settings — see below.

Main-window tabs (Chat / Files / Git / Rules)

• Chat — streaming output, collapsible tool steps, live token counts, context inspector, screenshots, voice input.
• Files — same as 🌐, following the current session's cwd.
• Git — git status (left) + git log (right). Click a changed file → side-by-side diff; click a commit → unified git show.
• Rules — edit KinetAios.md in the cwd (project-level rules, injected into the system prompt).

Pipeline (cross-engine orchestration)

Chain multiple stages, each with its own engine + prompt. The previous stage's output is prepended to the next stage's prompt. 2-minute per-stage timeout polling, abort on failure.

Session branching & handoff

• Branch from any turn — deep-copies turns/steps into a new session.
• Export/import sessions — serializes full session state (turns + history + engine + model + cwd) for cross-machine handoff. Export auto-redacts secrets (API keys, passwords → [REDACTED]).
• Cross-session references — link related sessions, visualized as a DAG.

Context management (Direct engine)

• Context inspector — view/edit each session's directHistory array (JSON editor).