Multi-agent orchestration · self-hosted node

Your AI agents.
Your node. Your keys.

A team of AI agents: an orchestrator that plans, specialists that execute, on a managed node you control. One flat price. Bring your own model keys. No credits, no execution meter, no surprise bills.

No credit meter Bring your own keys You own your data
Early access: Windows desktop app macOS & Linux next. By joining you'll get early-access emails. Unsubscribe anytime.
CodeSummon · cockpittoday $0.42 · month $7.10
Scoutworking Forgeworking Siftqueued Ninjaidle Atlasorchestrating▶ delegating 3 tasks…
orchestrator research coding data
Open by design

Bring your models. Plug in your tools.

No walled garden: use any model on your own key, and connect any tool through the open MCP standard.

Any model
OpenRouterAnthropicOpenAIGoogle GeminiMistralGroqDeepSeekLlama
Any tool · MCP
GitHub
n8n
Slack
Postgres
Google Drive
+ any MCP server
All product names and logos are trademarks of their respective owners, shown to indicate MCP compatibility.
Why CodeSummon

The credit meter is broken.

Every other agent platform bills you per credit or per execution. One busy week and the bill triples, with no warning. We took the opposite bet.

🩸 The metered crowd

  • Credits or executions you have to ration
  • The platform resells your tokens at a markup
  • A traffic spike becomes a surprise invoice
  • Your work and data live on their terms

🟠 CodeSummon

  • One flat monthly price, full stop
  • Bring your own key; pay the model provider at cost
  • Usage spikes never touch what you pay us
  • Your node, your keys, your knowledge base
Do it live

Give Atlas a job.

Pick a goal and watch the orchestrator plan it, delegate to the right specialists, and write what it learns back to the knowledge graph, all on your own key.

Pick a goal above, then hit “Summon the team” →
Atlasorchestrator
Scoutresearch
Forgecoding
Siftdata
$0.00 this run · your key
See it work

Watch a team think out loud.

Not a chat box. A live cockpit where the orchestrator delegates and the specialists work, visibly, in real time.

A knowledge base that builds itself

Every finished task writes a linked note back to your knowledge graph. The team reuses what it already knows instead of redoing work, and you can watch the connections grow.

🕸️ Auto-linked notes: semantic + wiki-style links between everything the agents learn.
🔎 Searchable & visual: explore the graph, or let agents query it before they research.
📤 Exports in one click: JSON, XLS, PDF. Zero tokens, zero lock-in.
CodeSummon · knowledge graph38 notes · 61 links
Pricing model Competitors BYO keys Unit economics Managed node

Delegation you can actually see

Ask Atlas a goal. It plans, then fans the work out to the right specialists (research, data, coding), each in its own window, with a live link back to the orchestrator while it runs.

🧭 Real planning: Atlas breaks a goal into steps and assigns each to an agent.
Live activity: animated links light up while an agent works; windows pop open on delegation.
📊 Honest cost HUD: today's and this month's spend, on your own key.
Atlas · orchestrator2 working
$ research international AI agencies, find best practices… ▶ planning · 3 steps → delegated to Scout · market scan tool: web_search · kb_write → delegated to Forge · draft outreach script tool: kb_search Sift · structure results ✓ done wrote 4 notes to the knowledge graph Delegated 3 tasks. They're running in the background.
What you get

A real team, not a single chatbot.

An orchestrator named Atlas plans the work and delegates to specialists. Everything is visible in a native desktop cockpit.

🧭

Orchestrator + fleet

Atlas breaks a goal into steps and hands each to the right specialist (research, data, or coding), then reports back.

🕸️

Self-growing knowledge graph

Every finished task writes back to a linked knowledge base, so the team reuses what it already knows.

🖥️

Native desktop cockpit

Floating agent windows, a live ops board, project workspaces. Watch your agents work in real time.

🥷

Approval-gated web research

Ninja, your scraper agent, only hits the live web when you arm it. Powerful research, on your terms.

📊

Live cost HUD

Today's and this month's model spend at a glance. It's your key, so the numbers are honest and yours.

📤

Export everything

Your knowledge graph and notes export to JSON, XLS and PDF in one click. Zero tokens, zero lock-in.

How it works

From prompt to a working team in seconds.

Install the cockpit

Grab the desktop app and sign in to your private managed node.

Windows in early access

Drop in your key

Paste your OpenRouter / Anthropic key once. We never bill against it.

Ask Atlas

Give the orchestrator a goal. It plans, delegates, and the specialists start working, visibly.

The graph grows

Results land in your content library and knowledge graph, ready to reuse next time.

Pricing

Flat price. Bring your own keys.

No credits. No execution meter. Your total cost stays lower because there's no markup on top of your own model bill. Annual billing takes ~20% off.

Solo

A single developer
$29/mo
1 seat · pooled node
  • Full agent fleet incl. Ninja
  • Unlimited projects
  • Knowledge graph + exports
Join waitlist
Most popular

Team

Growing teams
$79/mo
3 seats incl. · +$15/seat
  • Everything in Solo
  • Shared workspace + roles
  • Team knowledge base
Join waitlist

Business

Teams needing control
$299/mo
up to 15 seats · dedicated VPS
  • Everything in Team
  • SSO + audit log
  • Priority support
Join waitlist

Enterprise

Data-sensitive orgs
Custom
self-hosted or dedicated
  • SSO / SAML, on-prem
  • Security review + SLA
  • Dedicated support
Talk to us
🔑 Every plan is bring-your-own-keys Windows app at launch

No free tier. Early access is invite-only through the waitlist. Cancel anytime, no lock-in. Prices in USD.

Questions

Good to know

Which platforms are supported?+
Early access ships as a Windows desktop app. macOS and Linux builds are next on the roadmap. Join the waitlist and we'll tell you the moment they land.
What does “bring your own keys” mean?+
You add your own model provider key (OpenRouter, Anthropic, OpenAI…) once, in Settings. Your agents call the model directly with it, so you pay the provider at cost. We never resell tokens or bill against your key; we only charge the flat platform fee.
Why no credits or usage meter?+
Because we don't resell tokens. The biggest, most volatile cost in this space is model inference, and that's on your key, at cost. So we charge a simple flat price and never hit you with a surprise bill after a busy week.
Is there a free tier?+
No. A perpetual free plan is a heavy infrastructure cost for an early-stage team, so early access is invite-only through the waitlist and starts on a paid plan. There's no lock-in, and you can cancel anytime.
Who owns the data and the work?+
You do. Your projects, knowledge base and content live on your node, and everything exports to JSON, XLS and PDF in one click. No lock-in.

Summon your team.

Early access rolls out node by node. Drop your email and we'll open your spot as capacity comes online.

Early access: Windows desktop app · macOS & Linux next