Legion Intelligence
Also known as: Yurts, Yurts AI
Governed agentic AI for national security, delivered as role based Legion Packs that connect agents, models and data across existing systems under human approval, deployable from cloud to classified, air gapped and edge environments.
Legion Intelligence, formerly Yurts, builds governed agentic AI for national security and adjacent critical industries. Led by co-founder and chief executive Ben Van Roo, it serves the U.S. Department of War, the intelligence community, and the Department of Energy, with named customers including the XVIII Airborne Corps, the U.S. Air Force, and the Missile Defense Agency. Its premise is that humans command, agents operate, and every action is attributed.
Legion is sold as Legion Packs, role-based bundles that deploy a ready working environment on day one: Commander, Intelligence Officer, Operations Officer, and Maintenance Technician packs, each with Apps such as a SITREP Generator, Award Writer, Manual Assistant, and Work Order Scheduler. Underneath, the platform connects to more than 100 systems, including IBM Maximo, ServiceNow, SharePoint, Microsoft Teams, Jira, and Box, and supports MCP tools. Collections form a governed knowledge layer, so agents search an organization's documents and answer with citations traced to source, and no-code agentic workflows carry multi-step work across systems with people approving the steps that matter.
Governance runs through everything. Role- and attribute-based access control covers data, models, and agents, every action including an agent's is recorded in an append-only log showing who requested, approved, and changed the work, and an Eval Loop scores models, agents, and workflows against the organization's own measures, re-baselining each new model before it takes over a task. Model Command puts open-source, proprietary, fine-tuned, and government-furnished models behind one gateway. The company states FedRAMP High authorization for its cloud platform, SOC 2 Type II certification, CMMC Level 2, and authorization from IL2 through IL6.
Legion runs in cloud, on-premises, hybrid, air-gapped, and classified environments, and Centurion packages it with edge hardware for denied or disconnected settings, switching to locally hosted models when the network drops. Pricing is not published; it is set by network scope and user count, and the platform is awardable through the Tradewinds marketplace. For defense, government, and regulated operators that need governed agents inside sensitive or disconnected networks, it is built for exactly that; a small team wanting a self-serve builder will find it aimed at mission-scale deployments instead.
Vendor details
Canonical URL
https://www.legionintel.com
Category
Multi-agent platform
Subcategory
Secure agent orchestration for defense and enterprise
Funding status
Independent, based in San Francisco, formerly Yurts, led by co-founder and chief executive Ben Van Roo; deployed across the U.S. Department of War and Department of Energy.
Company status
independent
Use cases & customers
Primary use cases
Target customers
Deployment options
Integrations
Legion's Secure Connectors and data gateways provide allow listed integrations and entitled data paths into enterprise and mission systems including Microsoft SharePoint and Teams, Oracle, and defense systems like REMIS, CAMS, and Dataminr, with row and column level controls and end to end provenance. Its GenAI Fabric embeds and shares agents across applications and workspaces, and a multi model routing layer connects commercial and open source models.
In practice
Your analysts spend hours generating intelligence summaries and running repetitive staff workflows on classified networks where most AI tools cannot go. Legion runs governed agents inside those environments, cutting summary generation dramatically while attributing every action.
You want autonomous agents but cannot accept ungoverned actions on sensitive systems. Legion keeps humans in command, routes each task to an approved model, and enforces entitled data paths with end to end provenance and full audit trails.
Your mission needs AI at the edge in denied or disconnected conditions. Legion's Centurion edge system and air gapped deployment bring coordinated, secure agents to environments where cloud services simply are not available.
Sources & related URLs
Agentic Index coverage score
11.5 / 14 capabilities · 82%
| Integrations & Tool Calling | Full |
|---|---|
|
The platform connects to more than 100 systems out of the box, including IBM Maximo, ServiceNow, SharePoint, Microsoft Teams, Jira and Box, and supports MCP servers for agent tools, with agents acting across those systems under role and attribute based permissions. Sourcelegionintel.com/platformread 2026-09-27 |
|
| Workflow Orchestration | Full |
|
No code agentic workflows carry multi step work across systems end to end, with humans approving the steps that matter, and Apps such as SITREP Generator and Work Order Scheduler complete a task from start to finish inside a Space; how several agents coordinate is not detailed. Sourcelegionintel.com/llms.txtread 2026-09-27 |
|
| Knowledge Grounding & RAG | Full |
|
Collections are the governed knowledge layer behind every result, with permissions and provenance; agents search the organization's whole document ecosystem, answer with citations traced to source, and work only from the data a team scopes for them. Sourcelegionintel.com/llms.txtread 2026-09-27 |
|
| Human Oversight & Guardrails | Full |
|
Every step runs under human direction or approval according to policy the organization sets, approval workflows and guardrails govern agent actions, drafts get human review before anything ships, and the audit record shows who approved the work. Sourcelegionintel.com/llms.txtread 2026-09-27 |
|
| Security, Identity & Governance | Full |
|
Role based and attribute based access control govern data, models and agents, and the vendor states FedRAMP High authorization for its cloud platform, SOC 2 Type II certification, CMMC Level 2, NIST 800-53 alignment, HIPAA and GDPR attestations and authorization from IL2 through IL6. A trust center sits at trust.legionintel.com. Sourcelegionintel.com/llms.txtread 2026-09-27 |
|
| Observability & Auditability | Full |
|
Every action, including those taken by agents, is logged with full attribution in an append only record showing who requested, approved and changed the work and on whose behalf. Sourcelegionintel.com/platformread 2026-09-27 |
|
| Memory & State Persistence | Partial |
|
Spaces define what context is available to a team and the Operations Officer Pack is pitched as tracking activity without rebuilding context every shift, but no memory layer with a stated scope and lifetime that agents read and write is documented. Sourcelegionintel.com/llms.txtread 2026-09-27 |
|
| Deployment & Data Residency | Full |
|
Legion deploys in cloud, on premises, hybrid, air gapped and classified networks authorized IL2 through IL6, keeping customer data in the customer's environment, and Centurion runs the same Packs on four edge hardware configurations (a laptop, NVIDIA DGX Spark, HPE Edgeline and HPE ProLiant) with no cloud dependency. Sourcelegionintel.com/llms.txtread 2026-09-27 |
|
| Prebuilt Agents, Templates & Packs | Full |
|
Four role based Legion Packs ship ready, Commander, Intelligence Officer, Operations Officer and Maintenance Technician, each deploying a Space with agents, workflows, data connections and Apps such as SITREP Generator, Award Writer, Manual Assistant and Work Order Scheduler. Mission Kits, an earlier name, no longer appears on the site. Sourcelegionintel.com/llms.txtread 2026-09-27 |
|
| Triggers & Channel Coverage | Partial |
|
Work starts when people open a Pack's Apps in a Space, in cloud, on premises or on a Centurion edge node; no schedule, event or alert that starts an agent without a person is documented. Sourcelegionintel.com/llms.txtread 2026-09-27 |
|
| Model Flexibility & Routing | Full |
|
Model Command puts every reachable model, open source, proprietary, fine tuned or government furnished, behind one governed gateway where new models are added without code, connected deployments route to customer approved frontier models, and Centurion shifts to locally hosted open weight models when disconnected. Sourcelegionintel.com/platformread 2026-09-27 |
|
| APIs, SDKs & MCP Extensibility | Partial |
|
REST API extensibility is named, but no reference, coverage or access path for it is published. The MCP support described is for agent tools, with the platform calling out rather than being called. Sourcelegionintel.com/llms.txtread 2026-09-27 |
|
| Testing, Debugging & Optimization | Full |
|
The Eval Loop scores models, agents and workflows against the organization's own work using measures it defines, and every new model is re baselined against that work before it takes over a task. Sourcelegionintel.com/platformread 2026-09-27 |
|
| Browser & Computer Use | Not documented |
|
Agents act through connectors to 100+ systems and MCP tools; no browser or computer use capability is described. Sourcelegionintel.com/llms.txtread 2026-09-27 |
|
The Agentic Index coverage score grades every vendor Full, Partial or Not documented against the same 14 buyer facing capabilities, from public evidence only. Each capability links to how all vendors in the index score on it. How this evidence is graded
Pricing
Contact sales; pricing by network scope and user count
enterprise and government contracting, scaled to deployment, environments, and users
What is public
The pricing page states flexible, all inclusive pricing based on network scope and user count with no fees, and asks buyers to contact Legion. llms.txt adds that Legion is awardable through Tradewinds, supports direct award and sole source, and that Centurion is purchased as one contract with fixed pricing and a known bill of materials. No figures are published.
Billing mechanics
Presumed enterprise and government contracting scaled to deployment scope, environments, and users, with classified and edge deployments carrying additional accreditation and sustainment, though rates are not disclosed.
Cost watchouts
Deployments into classified, air gapped, or edge environments and government contracting carry integration, accreditation, and sustainment costs beyond a standard cloud subscription.
Variable cost rationale
Priced through enterprise and government contracts scaled to deployment scope, environments, and users, so cost grows with breadth of deployment and security requirements.
Additional watchouts
Confirm the contract vehicle, which environments are in scope, and the accreditation and sustainment obligations for classified, air gapped, or edge deployments.
Sales call required
Yes, required for paid access
Free / trial
Pilots and demos on request; no public free tier
Key ambiguities
No figure is published; price depends on network scope, user count and, for Centurion, the hardware configuration.
Related vendors
- CrewAI — Open source multi agent framework (Python, MIT) of role based Crews…
- Relevance AI — Platform for building and deploying a workforce of AI agents with a…
- Agent Zero — Open-source multi-agent framework that gives its agent a full Linux…
- Akka — Enterprise agentic AI platform that builds and runs multi agent…
- Artian — Enterprise multi agent AI platform for financial services: a Planner…
- Databricks Agent Bricks — Databricks' agent platform, now sold as Agent Bricks, for building,…
Alternatives to Legion Intelligence
The closest documented capability profiles to Legion Intelligence among multi-agent platforms tracked by Agentic Index, ordered by similarity on the same 14 point evidence the rankings use. No vendor pays for placement.
- Artian11.0 / 14A lighter documented profile than Legion Intelligence
- Databricks Agent Bricks13.0 / 14Fuller documented coverage on Memory & State Persistence and Triggers & Channel Coverage
- Dataiku13.0 / 14Fuller documented coverage on Memory & State Persistence and Triggers & Channel Coverage
- Relevance AI13.0 / 14Fuller documented coverage on Memory & State Persistence and Triggers & Channel Coverage
- Akka12.5 / 14Fuller documented coverage on Memory & State Persistence and Triggers & Channel Coverage
- Rasa11.5 / 14Fuller documented coverage on Triggers & Channel Coverage and APIs, SDKs & MCP ExtensibilityLegion Intelligence vs Rasa →
Similarity is computed from each vendor's Agentic Index coverage score evidence, axis by axis, not from the totals. How this evidence is graded