See how to govern AI agents & third-party risk with Obsidian. Live on Tuesdays. Register

40% of your agents are a critical risk. Read the CISO Playbook for Securing AI Agents

Announcing our Series D to fuel agentic productivity, securely

Secure your Microsoft Foundry apps and agents

Microsoft Foundry lets your teams build and deploy agents at scale. But knowing an agent is running is not the same as knowing what it can access, what it's doing, and whether it's creating risk. Obsidian extends enterprise governance to every Foundry agent, mapping real permissions, monitoring cross-application activity, and enforcing runtime guardrails before incidents occur.

Start Securing your Workflows Today

Why your teams are using Azure AI Foundry

Unlike prebuilt AI assistants available with Copilot, Azure Foundry lets users quickly build and deploy agents with custom AI models designed for the task at hand.

Automate complex workflows

Orchestrate multi‑step tasks like ticket triage, infra changes, CRM updates, and HR workflows.

Accelerate delivery

Use agents to write and review code, streamline processes, and improve customer experience.

Empower the workforce

Let non‑developers build workflows and agents without waiting on central engineering.

Foundry agents move fast, governance can't afford to lag

Unmonitored Foundry agents silently spread risk across every application they access and every data connection they make.

Sensitive data access

An over‑permissioned HR or finance agent can query classified internal knowledge across apps like SharePoint.

Code destruction risks

Agents wired into Azure DevOps or GitHub can overwrite or delete source code, pipelines, and infrastructure.

Data leaks at machine speeds

Business teams can expose CRM or ticket data as agents autonomously generate and share content externally.

Fragmented agent monitoring

Similar agent use cases can run in Vertex, Copilot, and Bedrock; without unified governance, fragmented policies and logs hide risky cross‑platform behavior.

Microsoft's native controls can’t secure every risk

Microsoft surfaces your agents but can’t show you every risk, leaving security teams blind to unauthorized tool calls, excessive permissions, and unsanctioned cross-service actions.

Secure your agents

Siloed visibility across tenants

No single view of which agents, MCP servers, and models are running across your tenants.

No single control plane

Native logs weren't built to capture risky tool calls and cross-service actions, especially from agents running outside Foundry on platforms like Claude.

Over-permissioned agents

Agent permissions are scattered across every app they touch. Without a unified view, you can't know your true exposure until something goes wrong.

Privilege escalation

Agents act on behalf of users but aren't always bound by the same limits. Without a full identity graph, you won't know when an agent quietly exceeds the access its user was granted.

Your single control plane for every Foundry agent you deploy

Map, monitor, and control every Foundry agent from a single governance layer.

Inventory every Foundry AI agent

Maintain a continuous system of record for every AI agent, including the MCP servers they invoke, the LLMs behind them, and the applications they connect to.
Shadow AI and auditability: Find unsanctioned agents including their connections and executions.
Track every agent: Map agents and know their blast radius no matter the platform they are built on.

Remove unsanctioned agent connections

Trace every Foundry agent’s connections and compare to approved apps and MCP servers to reduce the blast radius and prevent data exposure.
Review MCP connections: Remove unapproved agent connections to risky MCP servers.
Protect sensitive systems: Limit agent access to only approved systems.