[ Director ] 01/24
A conductor for the swarm
One Director routes intent into 25 specialist agents, splits work in parallel, gathers results, and decides what to do next.
engine.dispatch · 25 specialists
v0.2.27 · stable
macOS · Linux · Windows · Raspberry Pi
The operating system
for agent swarms.
25 specialist agents. 511 workflow integrations. 19,000+ skills. Voice chat, image gen, cloud sync, OAuth, encrypted vault — every feature for every customer. One-click installers for six platforms. License-gated, fully local. No SaaS middleman.
$100 per key / month · bulk discount from 10 keys (5%) → 30 keys (25%) · 35+ enterprise
- Integrations
- 511+
- Skills
- 19K+
- Agents
- 25+
- Platforms
- 6
[ swarm.viewport ]
live
director · idle 25 nodes / 187 edges
DISPATCH director → researcher · 12ms MCP github → fetched 47 issues VAULT secrets refreshed · 42 keys CRON 0 9 * * 1 → digest run · 2.3s DISCORD reply · #ops WORKFLOW invoice-export.v3 · ok SKILL crawl-news.v7 · 0.85s EMBED 1283 chunks · 312KB PROVIDER ollama-cloud · 18 reqs/min iMESSAGE inbound · alex TASK created · #q3-pricing POWER-UP webull · session ok CHANNEL discord/general · 03 events AGENT critic ↘ writer · feedback FILE inbox/contract.pdf → vault CRON */15 * * * * → market-pulse DISPATCH director → researcher · 12ms MCP github → fetched 47 issues VAULT secrets refreshed · 42 keys CRON 0 9 * * 1 → digest run · 2.3s DISCORD reply · #ops WORKFLOW invoice-export.v3 · ok SKILL crawl-news.v7 · 0.85s EMBED 1283 chunks · 312KB PROVIDER ollama-cloud · 18 reqs/min iMESSAGE inbound · alex TASK created · #q3-pricing POWER-UP webull · session ok CHANNEL discord/general · 03 events AGENT critic ↘ writer · feedback FILE inbox/contract.pdf → vault CRON */15 * * * * → market-pulse
[ 1 · manifesto ]
The opposite
of "AI cloud".
Built for people who don't want to rent their intelligence — and who'd rather know exactly which process is running which model on which CPU.
01
Local by default.
Models run on your hardware. Data stays in your filesystem. Nothing is forwarded unless you say so. No telemetry. No fine-print analytics.
02
Composable, not closed.
Every agent is a swappable module. Skills are scripts. Channels are adapters. MCP servers plug into the same bus. You ship the swarm you need.
03
Long-lived, not one-shot.
Cron, queues, and durable workflows make agents wait, retry, and hand work back and forth. Your AI is a daemon, not a chat box.
[ 2 · capabilities ]
Everything an
operating system for agents needs.
modules / 24
runtime / v0.2.27
status / ok
runtime / v0.2.27
status / ok
[ Workflows ] 02/24
511 integrations, drag-and-drop
Visual graph editor with n8n parity plus AI/vector/RAG nodes. Sub-workflows, error workflows, per-flow concurrency caps, real-time exec viz. Export PNG/SVG.
511 nodes · git history · debugger
[ Skills ] 03/24
19,000+ plug-and-play skills
Searchable shelf. Install with one click. From web crawl to invoice gen to market scrape — every skill an agent can call.
19K shelf · signed · forkable
[ Voice ] 04/24
Talk back-and-forth, naturally
Free Microsoft Edge neural voices (Aria, Jenny, Ava). Whisper STT via Groq or OpenAI. Continuous conversation mode with silence detection.
Edge TTS · Whisper · convo loop
[ Image + vision ] 05/24
Generate, see, describe
DALL·E 3, FLUX schnell, Stable Diffusion 3.5 nodes. GPT-4o vision lets agents describe any image URL or paste. Drop a screenshot, ask anything.
DALL·E · FLUX · SD3.5 · vision
[ Avatar ] 06/24
Talking-head video on demand
Pipe agent text through D-ID Talks for a video clip of a face speaking it. Stock presenters or upload your own headshot.
D-ID Talks · 4 stock faces
[ Vault ] 07/24
AES-GCM encrypted, local-only
Provider keys + OAuth tokens sealed with a passphrase you set (PBKDF2 200k iters). Lock/unlock from the UI. Nothing leaks to logs.
AES-GCM · PBKDF2 · scoped
[ OAuth broker ] 08/24
Login flows, not key paste
Built-in flows for Google, GitHub, Slack, Notion, Microsoft. Token refresh handled automatically. Add custom OAuth 2.0 providers.
5 built-ins · auto refresh
[ Memory ] 09/24
Vector-indexed, agent-shared
Notes + wiki-links + transcripts. Auto-embedded with text-embedding-3-small for semantic search across thousands of entries. Plain markdown on disk.
embeddings · obsidian · graph
[ MCP servers ] 10/24
Any tool the model knows
Hook in any Model Context Protocol server. GitHub, Linear, Gmail, Drive, Stripe, Postgres. Connect once, reuse forever.
discover · install · pin
[ Channels ] 11/24
Reach you where you live
Native iMessage, Discord, email, Slack. Agents reply, react, ping push notifications when long jobs finish. Always-on or scheduled.
macOS · DMs · groups
[ Cost router ] 12/24
Cheapest model that meets the bar
Hand-curated price table for 17 frontier→local models, tier 1–5. Quality threshold + spend cap → cost router picks the right tool every call.
17 models · tier 1–5
[ Live ops ] 13/24
See every tick of every agent
Real-time exec visualization paints workflow nodes blue while running, emerald on success, red on error. Director heartbeat, costs, latencies.
tracing · viz · cost
[ Auth + RBAC ] 14/24
Bring your team
Optional multi-user mode with admin/editor/viewer roles. scrypt password hashing, 7-day session tokens, per-user audit trails.
scrypt · sessions · roles
[ Cloud sync ] 15/24
Your snapshots, your bucket
Push config + workflows + memory to any S3-compatible bucket (AWS, R2, B2) or WebDAV (Nextcloud). End-to-end safe — your bucket, your keys.
S3 · WebDAV · ts-conflict
[ Environments ] 16/24
Dev / staging / prod scopes
Tag credentials with env scopes. Switch dev → prod with one dropdown — every key lookup snaps to the live scope. No accidental prod hits.
scoped · tagged · per-machine
[ Audit log ] 17/24
Every action, queryable
Append-only JSONL of every workflow run, git commit, login, admin action. Filter, search, export. WCAG-compliant viewer page.
JSONL · search · 5s tail
[ Eval harness ] 18/24
Catch prompt drift early
Define goldens (contains / equals / regex / LLM-judge), run them against any agent, watch pass-rate over time. Results stream to JSONL.
goldens · judge · trends
[ Cmd-K palette ] 19/24
Jump anywhere instantly
Press ⌘K — search across pages, agents, workflows, memory notes. Keyboard-first, deep-links to the right view.
⌘K · ↑↓ · enter
[ CLI + SDKs ] 20/24
Script everything from the terminal
agx CLI: list, run, chat, audit, eval, tts, sync. Plus single-file Python and TypeScript SDKs (zero deps). Drop into any project.
agx · python · typescript
[ Installers ] 21/24
1-click install, six platforms
Native DMG + PKG (Apple Silicon + Intel), AppImage (x86_64 + ARM64 for Pi 4/5), NSIS .exe (x64 + ARM64). Signed, auto-updating, rollback button.
Mac · Linux · Win · Pi
[ Safety ] 22/24
Security review on tap
Built-in safety-check skill audits any code or workflow before deploy. Flags secrets-in-code, dangerous shell, missing input validation.
skill · pre-deploy · OWASP
[ i18n + a11y ] 23/24
English, Spanish, accessible
Locale switcher with en/es bundles. Visible focus rings, skip-to-main link, prefers-reduced-motion. Add your own bundle in one file.
en · es · ARIA · keyboard
[ License-gated ] 24/24
Your runtime, your tokens, no markup
Every install needs a valid key — same key, same features for everyone. Bring your own provider keys, run local Ollama, never pay token markup.
$100 / key · BYO tokens
[ 3 · architecture ]
Five layers,
zero cloud hops.
Every layer is on your machine. Outbound only happens when you call a model provider — and you choose which calls leave the box.
Architecture deep-dive →- L1
UI · React + Vite
Dashboard, Chat, Live, Memory graph, Workflows. Loads from disk, runs in browser at localhost:8858.
:8858 - L2
Sidecar · Node
Long-lived host process. Routing, MCP host, channels, queues, cron, license. Streams SSE end-to-end.
engine - L3
Engine · runtime
Director, swarm scheduler, persona memory, skills runner, providers, embed pipeline.
core - L4
Vault · disk
~/.agentiqflow stores secrets (sealed), sessions, tasks, workflows, and your obsidian-style memory vault.
fs - L5
Models · pluggable
Anthropic, OpenAI, Gemini, NVIDIA NIM, OpenRouter, Ollama Cloud, local Ollama. Swap per agent.
providers - ↑Boundary · network egress only on your callopt-in
[ 4 · providers ]
42 providers.
One bus.
Configure once. Route per-agent. Mix frontier with open-weights, fast with cheap, hosted with local. Failover and budget caps are built in.
Routing
per agent
Caching
prompt+context
Failover
automatic
Budgets
$ + tokens
Anthropic
Claude Opus / Sonnet / Haiku
01/42
OpenAI
GPT-5 family
02/42
Google
Gemini Pro / Flash
03/42
NVIDIA NIM
Hosted + on-prem GPUs
04/42
Mistral
Mistral Large / Codestral
05/42
Cohere
Command R+
06/42
Groq
Llama / Qwen at speed
07/42
Together
Open-weights at scale
08/42
Fireworks
Tuned + LoRA serving
09/42
OpenRouter
Unified relay · 200+ models
10/42
Ollama Cloud
Hosted open-weights
11/42
Ollama local
Run on your GPU
12/42
vLLM
Self-hosted inference
13/42
Replicate
Image / video / audio
14/42
DeepSeek
DeepSeek V3 / R1
15/42
xAI
Grok 3
16/42
Perplexity
Sonar online models
17/42
Qwen
Qwen 3 family
18/42
+ 24 more · see /models
[ 5 · pricing ]
$100 per key, per month.
Every feature.
One license, one monthly price, every capability unlocked — voice, image gen, cloud sync, OAuth, RBAC, all 511 integrations, all 19,000 skills. Stack keys for multiple machines or your team — bulk discount kicks in at 10 keys and grows 5% every five keys after that.
[ 1 key ]
$100
/ key / mo
$100/mo total
Buy 1 license [ 10 keys ]
$95
/ key / mo
$950/mo total · 5% off
Buy 10 keys [ 15 keys ]
$90
/ key / mo
$1350/mo total · 10% off
Buy 15 keys [ 20 keys ]
$85
/ key / mo
$1700/mo total · 15% off
Buy 20 keys [ 25 keys ]
$80
/ key / mo
$2000/mo total · 20% off
Buy 25 keys
Best value
[ 30 keys ]
$75
/ key / mo
$2250/mo total · 25% off
Buy 30 keys 35+ keys?
Buying 35 or more keys? Contact us for a custom enterprise quote with deeper volume pricing, SSO, on-prem update server, and a named support engineer.
[ included with every key ]
Every feature, every key. No upsells.
- 511+ workflow integrations (n8n parity, plus AI/vector/RAG)
- 19,000+ skills library — install with one click
- 25 specialist agents (Director, Coder, Trader, Writer, …)
- Voice chat — talk to agents, agents talk back
- Microsoft Edge neural TTS — free Aria/Jenny/Ava voices
- Whisper STT (Groq + OpenAI), browser fallback
- Image generation — DALL·E 3, FLUX, Stable Diffusion 3.5
- Vision input — drop an image, ask anything
- Avatar agent — talking-head video via D-ID
- Encrypted vault — AES-GCM + PBKDF2 200k iterations
- OAuth broker — Google, GitHub, Slack, Notion, Microsoft
- Cost router — auto-pick cheapest model meeting quality bar
- Memory with embeddings — semantic search across notes
- Workflow versioning — git history + restore in UI
- Workflow debugger — pinned inputs + replay-from-node
- Sub-workflow nodes — compose complex flows from small ones
- Error workflow — auto-fire alert/retry flow on failure
- Per-workflow concurrency + rate limits
- Webhook HMAC auth (GitHub / Slack / Sentry signatures)
- Real-time execution viz — live node-by-node highlights
- Export workflows as PNG / SVG / JSON
- Templates marketplace — community + custom registries
- Audit log — every action recorded, queryable in UI
- Multi-user auth + RBAC (admin / editor / viewer)
- Cloud sync — S3-compatible or WebDAV, end-to-end safe
- Dev / staging / prod environment scopes
- Cmd-K palette — search agents, workflows, memory, files
- CLI (agx) + Python SDK + TypeScript SDK
- i18n — English + Spanish included
- Accessibility — keyboard nav, ARIA, reduced-motion
- One-click installers — DMG, PKG, AppImage, NSIS .exe
- Native builds — Mac arm64+Intel, Linux x64+ARM64, Win x64+ARM64
- Raspberry Pi 4/5 installer with systemd unit
- MCP server hub — install + manage from UI
- Channels — Discord, iMessage, Slack, email integrations
- Memory + obsidian-vault sync
- Cron — recurring schedules with full execution history
- Live dashboard — agent activity, costs, heartbeat
- Backup / restore — JSON snapshot + cloud sync
- Auto-update — staged rollouts, rollback button
- License-gated — your runtime, your tokens, no markup
- Safety check — security review skill on demand
License works on
1 device per key. Need more? Stack keys — bulk pricing kicks in at 3+. Master keys (unlimited devices) available on request.
Token spend
You bring your own provider keys (or run local Ollama). We never markup a single token. The cost router auto-picks the cheapest model meeting your quality bar.
All sales final
No refunds. You get every feature, every update, instant key delivery. Buy with intent — try the docs + demo before you commit.
[ 6 · download ]
Release
0.2.27
Native installers for 8 platform/arch combos — Mac (Apple Silicon + Intel), Linux (x86_64 + ARM64), Windows (x64 + ARM64), Raspberry Pi (4 / 5), plus a universal Node tarball. Signed builds, SHA-256 published, auto-update on stable.
[ release.manifest · v0.2.27 ]
verified
| Target | Format | Get |
|---|---|---|
| macOS Apple Silicon · arm64 | DMG | Download → |
| macOS Apple Silicon · arm64 | PKG | Download → |
| macOS Intel · x86_64 | DMG | Download → |
| macOS Intel · x86_64 | PKG | Download → |
| Linux x86_64 | AppImage | Download → |
| Linux ARM64 · aarch64 | AppImage | Download → |
| Windows x64 | EXE | Download → |
| Windows ARM64 | EXE | Download → |
| Raspberry Pi Pi 4 / Pi 5 · 64-bit Pi OS | SHELL | Download → |
| Any · Node 20+ Universal · all arches | TAR.GZ | Download → |
Installers + tarball gated by license at runtime. Direct GitHub links above.
Enter license key →
[ 7 · faq ]
Questions
worth answering.
Is my data ever sent to AgentiqFlow?
[ open ] [ close ]
No. The runtime is local. The only network calls are: model providers you configure (using your keys), MCP servers you install, and our update / license server. The update server only sees your license key + device ID + app version.
01/11
Do I need a GPU?
[ open ] [ close ]
No. Most agents work fine through hosted providers. You only need a GPU if you want to run local Ollama models or local video / image diffusion.
02/11
How are model costs handled?
[ open ] [ close ]
We never markup tokens. You bring your own API keys (or run local). We optionally offer per-agent budget caps, daily ceilings, and prompt-cache analytics inside the app.
03/11
What is a Skill vs an MCP server vs a Power-up?
[ open ] [ close ]
Skills are local scripts your agents can call (Python or Node). MCP servers are tool servers that follow the Model Context Protocol — agents discover them and use their tools. Power-ups are first-party native integrations (Webull, iMessage, Drive) that wrap platform APIs and store credentials in your vault.
04/11
Can teams use this together?
[ open ] [ close ]
Yes — multi-user auth + RBAC ships in every key (admin / editor / viewer roles, scrypt password hashing, 7-day session tokens). Buy a key per teammate; the bulk discount kicks in at 3+ keys (10% off), 5+ (20%), 10+ (30%). Cloud sync (S3 / WebDAV) keeps workflows aligned across machines.
05/11
Is the source available?
[ open ] [ close ]
Core runtime is MIT-licensed and on GitHub. Premium components (channels, cloud sync, marketplace) are source-available with paid license.
06/11
What about offline mode?
[ open ] [ close ]
Yes. With local Ollama models, you can run a fully offline swarm. The license server tolerates 30 days offline without re-check.
07/11
Is there a free tier or trial?
[ open ] [ close ]
No free tier — every download, install, and runtime invocation needs a valid, paid license. Pricing is flat: $100 per key, every feature included for everyone. Bulk: 3 keys $90 each, 5 keys $80 each, 10+ keys $70 each. All sales final — no refunds. Try the docs + demo before you buy.
08/11
What's actually included in a key?
[ open ] [ close ]
Everything. All 511 workflow integrations, all 19,000+ skills, voice chat (Edge TTS + Whisper), image gen (DALL·E / FLUX / SD3.5), vision, avatar agent (D-ID), encrypted vault, OAuth broker, cloud sync, RBAC, debugger, eval harness, CLI, SDKs, i18n. No tier locks, no upsells.
09/11
How many devices per key?
[ open ] [ close ]
One device per key. For multiple machines, stack keys — bulk discounts kick in at 3+. Master keys (unlimited devices) available on request — email enterprise@agentiqflow.ai. Bulk-pack keys are independent — each teammate gets their own key for their own device.
10/11
What happens if my subscription lapses?
[ open ] [ close ]
If a payment fails or you cancel, your license flips to past_due (Stripe dunning) then revoked. The local sidecar checks license status with the worker on a 5-minute cache; once revoked, agent dispatch is gated and the dashboard prompts you to reactivate billing.
11/11
[ ready ]
Stop renting your
intelligence.
Install in 90 seconds. Bring your keys. Wire up your channels. Watch your swarm do work while you do other things.