The Pulse
FlowForma launches AI agent monitoring and control hub
FlowForma Agent Trust is available to organizations that want to monitor AI agents and route their work for human review. The company says the tool connects agents from different platforms and records oversight decisions.

FlowForma has launched Agent Trust, a hub it says lets organizations monitor AI agents, pause them or send their work for human review. FlowForma says the hub, launched on November 18, connects agents from different platforms and records oversight decisions.
Source prnewswire.com
The product is available now. It is intended for operations, IT, AI, risk and compliance teams that need to see how agents are performing alongside people and business processes.
Connecting agents and tracking work
Prebuilt connectors cover platforms such as Microsoft Copilot Studio, Power Automate and n8n; the release says custom-built agents and agents from other platforms can also connect. The company says the system discovers agents, including ones that were not registered, and gives each an identity and an owner.
Organizations can choose policies for an agent, including approval limits and requirements for human review. Once an agent is connected, the product is designed to monitor its runs and associate them with the business cases they serve.
The agent view tracks how much work agents handle, complete independently and hand over to people. A trust view plots an agent's autonomy against how often people correct its output. The system also separates handoffs required by an organization's rules from those involving work an agent could not complete, according to the product page.
Controls and oversight records
The release says users can pause an agent or send its work for human review in one click, then confirm that the change took effect. The product page also describes an option to slow an agent down and says recommendations can show the expected cost and fixes associated with an action.
Agent Trust records pauses, releases, overrides and policy changes as they occur. The company presents that record as evidence of human oversight for regulated organizations, citing the EU AI Act as an example.
Alongside agent monitoring, the product page describes views for processes and people. These include tracking work in progress or overdue, locating bottlenecks, viewing team availability and planning for absences.
AI use and data handling
FlowForma's AI disclosure says Agent Trust may use AI to summarize and classify agent activity, detect agents that may be making errors using human corrections, and generate risk scores, alerts and suggested controls. Its typical outputs include performance reports and suggested reviews or pauses.
The disclosure says AI output is decision support and can be incomplete, inaccurate, outdated or wrong in context. It assigns customers responsibility for reviewing outputs, testing agents before live use and making final decisions.
FlowForma products use Microsoft Azure OpenAI Service. For Agent Trust, stored data is held in Ireland, the Netherlands and Sweden, with the latter used for AI service data. Individual AI requests may be processed in any Azure region where the model is available, so processing is not limited to the European Economic Area, the disclosure says.
FlowForma will demonstrate the product in a webinar on November 24 at 3 p.m. GMT, or 10 a.m. EST. The session is intended for operations, IT, AI, risk, compliance and audit leaders.
Sources
Related stories
- Microsoft and partners start enterprise AI transparency initiative
- Microsoft introduces enterprise policy controls for AI agents
- Microsoft makes Copilot Home generally available as Autopilot rolls out
- Workday plans to buy Honeydew for governed AI queries
- Microsoft and Salesforce bring sales workflows to Copilot in preview