Moxo
Also known as: Moxo Agent Foundry, Moxo AI, moxo.com
Moxo is a workflow platform for business processes that run between companies, their clients and their partners, and its Agent Foundry puts AI agents into those processes. Prebuilt agents such as a compliance reviewer read submissions, validate documents and chase overdue items, then hand a prepared decision to a named person. Supervisor agents check work and escalate, every run is traced, and Moxo deploys in its cloud, a private cloud or on premises.
Moxo sells an AI workflow platform for business operations that run between a company, its clients and its partners, where people, AI agents and systems all have a part in the same process. Branded client portals bring clients and partners into the work, and forward deployed engineers help configure processes to each customer's systems and policies. Customers include Savills, Procore, FIS, Scotiabank, Citi, Emirates Bank, ChenMed, Zeta Global, Palmetto and CD Baby, and Peninsula Visa cut its processing time in half.
Agent Foundry is where customers build, run and supervise AI agents inside those processes. A library of prebuilt operations agents, such as a compliance reviewer and an analysis agent, reads submissions, validates documents, chases overdue items and passes a prepared decision to the team. Teams ground agents in their own playbooks, policies, knowledge and price lists, chain agents so one's output feeds the next without code, and run agents built on OpenAI, Anthropic or their own stack beside Moxo's. Agents produce branded PDFs, Excel models, Word documents and decks, and Moxo AI reads large volumes of documents, pre fills forms and flags patterns across submissions.
The operating principle is that humans decide and AI executes. Each agent has a defined scope and a named person who receives anything outside it, supervisor agents check work on key steps and escalate issues before they reach a client, and people make approvals and exception calls.
Every run is traced from what the AI saw to what it recommended and what the person decided, and agents are judged on outcomes such as whether a reviewer reopens approved work. Each run also writes to a context graph of decisions and patterns that stays inside the customer's organization, so agents learn recurring exceptions and stalls. Model routing picks a model per task, and cost caps limit spending on test flows.
Moxo sells through demos and trial requests, with no public price list. It holds SOC 2 Type II and SOC 3 reports, offers granular role based access with customizable privileges and keeps an audit trail of seven years or more for regulated industries. It deploys as multi tenant cloud, in a private cloud or on premises. Moxo suits banks, property firms and other organizations with document heavy processes that involve clients and partners, while teams wanting a general purpose agent builder or public developer documentation will find an operations platform first.
Vendor details
Canonical URL
https://www.moxo.com/
Category
Enterprise operations agent
Subcategory
Agents inside multiparty business process orchestration
Company status
independent
Use cases & customers
Primary use cases
Target customers
Deployment options
Integrations
Moxo APIs, webhooks and Zapier connect processes to a customer's systems, branded client portals bring outside parties in, and agents built on OpenAI, Anthropic or a customer's own stack can run alongside Moxo's agents.
In practice
A bank's onboarding process has a compliance reviewer agent check uploaded documents and pass a prepared decision to an analyst.
A property firm's agent chases tenants for overdue paperwork through the client portal and escalates stalled cases to a named manager.
A payments company chains an extraction agent and an analysis agent to turn merchant submissions into an Excel risk model for review.
Sources & related URLs
Related / legacy domains
Agentic Index coverage score
10.5 / 14 capabilities · 75%
| Integrations & Tool Calling | Partial |
|---|---|
|
Processes connect to a customer's systems through webhooks, Moxo APIs and Zapier, which push data in and pull it out, and implementation teams configure those connections to each customer's systems. Agents produce documents, pre fill forms and send follow ups inside the process. Moxo names no integration catalog or tool actions that its agents take directly in outside business applications. Sourcemoxo.com/integrationsread 2026-11-20 |
|
| Workflow Orchestration | Full |
|
Moxo orchestrates processes that involve people, AI agents and systems across teams and organizations. In Agent Foundry, one agent's output feeds the next agent's input in multi agent workflows built without code, supervisor agents check work on key steps, and anything outside an agent's scope moves to a named person, so a process can run from intake to decision. Sourcemoxo.com/ai-agentsread 2026-11-20 |
|
| Knowledge Grounding & RAG | Full |
|
Teams upload policies, playbooks, knowledge articles and price lists as agent knowledge, so agents work from the organization's own sources rather than generic model behavior. Agents are grounded in the same process context as the teams they work beside, including the submissions and documents moving through each workflow, and Moxo AI reads large volumes of documents to flag patterns. Sourcemoxo.com/ai-agentsread 2026-11-20 |
|
| Human Oversight & Guardrails | Full |
|
People make the approvals, exception calls and escalation decisions while agents prepare the work. Each agent has a defined scope and a named person who receives anything outside it, and supervisor agents on key steps check work, flag anomalies and escalate before an issue reaches a client. Agents act only within the permissions an administrator sets. Sourcemoxo.com/airead 2026-11-20 |
|
| Security, Identity & Governance | Full |
|
Granular role based access controls with customizable user privileges govern who can see and do what, alongside detailed activity tracking. Moxo holds SOC 2 Type II and SOC 3 reports, states GDPR compliance, follows Cloud Security Alliance practices and encrypts data in transit and at rest. Customer context data is never used to train other customers' instances. Sourcemoxo.com/securityread 2026-11-20 |
|
| Observability & Auditability | Full |
|
Every run is traced from what the AI saw to what it recommended and what the person decided, so a reviewer can see why a case moved the way it did. Every action by people and agents lands in an audit trail kept for seven years or more for regulated industries, and supervisor agents flag anomalies live while the work is still underway, before a client sees it. Sourcemoxo.com/airead 2026-11-20 |
|
| Memory & State Persistence | Partial |
|
Each run writes to a context graph that records decisions and patterns, so agents learn the recurring exceptions and stalls in a customer's processes over time. That memory stays inside the customer's organization, is never pooled with other customers and is never used for training. Moxo gives it no stated retention period or way to clear what agents have learned. Sourcemoxo.com/ai-agentsread 2026-11-20 |
|
| Deployment & Data Residency | Full |
|
Moxo deploys as a multi tenant cloud service, in a private cloud in the customer's own environment or on premises, so regulated customers can keep processes, documents and client records inside their own perimeter. Context data used by agents stays within that perimeter too. Moxo names no specific hosting regions for the multi tenant version of its cloud. Sourcemoxo.com/securityread 2026-11-20 |
|
| Prebuilt Agents / Templates / Packs | Full |
|
Agent Foundry includes a library of prebuilt operations agents modeled on strong operations teams, such as a compliance reviewer and an analysis agent, each with its own task and adjustable to a customer's workflows. Agents produce branded PDFs, Excel models, Word documents and decks. Moxo names only these examples rather than the full library of agents. Sourcemoxo.com/ai-agentsread 2026-11-20 |
|
| Triggers & Channel Coverage | Partial |
|
Work reaches agents as process steps, such as a submission arriving through a branded client portal or a document waiting for review, and agents chase items that sit overdue. Webhooks and Zapier carry events between Moxo and other systems. Moxo names no schedule, inbound email or event trigger types for starting agents beyond the processes they belong to. Sourcemoxo.com/ai-agentsread 2026-11-20 |
|
| Model Flexibility & Routing | Partial |
|
Model routing picks a model for each task automatically, using higher quality models where accuracy matters and faster ones elsewhere, without naming them or letting customers choose. Teams can bring their own agents built on OpenAI, Anthropic or another stack to run beside Moxo's, which keeps their choice for those agents but not for Moxo's own agents. Sourcemoxo.com/ai-agentsread 2026-11-20 |
|
| APIs / SDKs / MCP Extensibility | Partial |
|
Moxo APIs, webhooks and Zapier connect the platform to other systems, and outside agents can run inside Moxo processes alongside its own. No public API reference, SDK or developer portal sits behind those APIs, and Moxo describes no MCP server or agent to agent interface, so building on the platform runs through its own team and implementation services. Sourcemoxo.com/integrationsread 2026-11-20 |
|
| Testing, Debugging & Optimization | Full |
|
Agents are evaluated on outcomes after they go live, such as whether a reviewer reopens work an agent approved, which measures the agent's decisions against what people later conclude. Test flows run under cost caps that set hard spending limits for experiments before rollout. Moxo names no scored test set that runs before a change to an agent ships. Sourcemoxo.com/airead 2026-11-20 |
|
| Browser / Computer-use | Not documented |
|
Agents work on documents, forms and submissions inside Moxo processes and portals, and data moves to other systems through webhooks, APIs and Zapier. Client portals are interfaces for people outside the company to use and sign in to. Moxo describes no agent that drives a browser, a desktop session or a remote computer to finish work in another application. Sourcemoxo.com/ai-agentsread 2026-11-20 |
|
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
Quote only, after a demo or trial request
Enterprise subscription with deployment and implementation services
What is public
Moxo sells its platform, Agent Foundry and client portals through a demo or a free trial request, with prices quoted by sales.
Billing mechanics
Each agreement sets the deployment, whether multi tenant cloud, private cloud or on premises, along with users, processes and agents. Cost caps set hard spending limits on agent test flows.
Sales call required
Yes, required for paid access
Free / trial
Free trial on request
Commercial notes
Forward deployed engineers and implementation services help configure processes and integrations to each customer's systems.
Missing data
Moxo quotes every price after a sales conversation, so plan contents and AI usage charges sit in the order.
Related vendors
- 11th Estate — Agentic platform whose AI engine scans markets, matches an…
- Abacum — FP&A platform for growing companies whose AI Modeler builds, edits…
- Accend — Accend builds human in the loop AI agents for commercial credit…
- Adopt AI — Finance agents that categorize, reconcile and close the books inside…
- Aera Technology — Decision intelligence platform where an always on agent executes…
- Akro AI — On premise operational intelligence that automates document heavy…