AI action control
Apply policies to agent requests across remote and local MCP servers, APIs, CLIs, and agent-generated code. Access can be scoped by user and user group, with requests checked before they reach the target.
Obot
Freemium
Obot is an open-source enterprise AI control plane and MCP gateway for deploying, governing, and auditing AI agents, MCP servers, and Skills. It helps organizations centralize access policies, discovery, authentication, and activity records across AI tooling.
什么是 Obot？
Obot is an enterprise AI control plane and MCP gateway for organizations adopting AI agents, MCP servers, and Skills. It provides a central place to connect AI tooling to company data, apply access policies, and review activity across integrations. Obot is available as open-source software for self-hosting and also as a hosted service.
The control plane covers the lifecycle from deployment and identity setup through catalog management, policy enforcement, discovery, and audit. Teams can launch Obot in the cloud or in their own environment, connect an identity provider, and give users governed access to approved tools through gateway URLs and AI clients.
Obot also addresses unmanaged AI usage. Its Sentry component scans user-machine configurations for agents, MCP servers, Skills, tools, and plugins, while the platform records MCP activity and exposes token usage by agent, user, and group. This gives platform, security, and compliance teams a shared view of AI tooling and its activity.
Obot 能做什么？
MCP and Skills catalog
Provide a central registry where teams can discover, install, request, and manage MCP servers and Skills. Administrators can review, approve, and publish catalog entries.
Audit logs
Record requests and responses across MCP servers in a consolidated, exportable audit trail. Teams can query the records, analyze activity, and generate reports.
Shadow AI discovery
Obot Sentry scans user-machine configurations to inventory unmanaged agents, MCP servers, Skills, tools, and plugins. Administrators can identify and disable blacklisted MCPs and Skills from the dashboard.
Authentication and deployment
Launch Obot in the cloud or self-host it with Docker or Kubernetes, then connect an identity provider such as GitHub, Google, Okta, or Microsoft Entra. The gateway places SSO and OIDC in front of connections.
Token usage visibility
Show model consumption by agent, user, and group, and apply policies governing which users and groups can call particular models.
使用场景
“Govern an internal AI tool registry”
Platform or security teams can host internal MCP servers, select servers from the catalog, and publish approved tools and Skills for organization-wide discovery.
“Investigate unmanaged AI activity”
Security teams can scan device configurations for shadow agents and MCPs, review alerts and collected logs, and disable tools that fall outside organizational policy.
“Control agent access to business systems”
Administrators can define which users and teams may discover or call specific MCP servers and Skills, helping constrain agent access to connected data and actions.
“Support compliance and operational review”
Security and compliance teams can inspect a centralized record of MCP requests and responses, export the audit data, and use it for reporting.
“Manage model consumption”
Platform teams can review token usage by agent, user, or group and restrict model access as they establish cost controls for a growing model stack.
常见问题
How can Obot be deployed?
Obot can be launched as a hosted cloud service or self-hosted in an organization’s own cloud or on-premises environment. The site also references Docker and Kubernetes deployment options.
Which identity providers are supported?
The documented setup examples include GitHub, Google, Okta, and Microsoft Entra. Obot also describes SSO and OIDC support for authenticated connections.
What does Obot log?
Obot records requests and responses across MCP servers, including MCP and tool activity. The resulting audit trail can be reviewed, queried, analyzed, and exported.
Can Obot detect unmanaged AI tools?
Yes. Obot Sentry scans AI client configurations across user machines and reports unmanaged agents, MCP servers, Skills, tools, and plugins. The dashboard can disable blacklisted MCPs and Skills.
Is Obot open source?
The product site describes Obot as 100% open source, with options to inspect the codebase, self-host it, and create or connect custom MCP servers.
快速信息
- Product type
- Enterprise AI control plane and MCP gateway
- Deployment
- Hosted cloud, self-hosted cloud, or on-premises
- Authentication
- SSO and OIDC; examples include GitHub, Google, Okta, and Microsoft Entra
- Primary workflows
- AI access control, MCP and Skills discovery, shadow AI scanning, audit, and token usage monitoring
- Open-source status
- Open-source software with self-hosting and code-audit options
- Catalog coverage
- The MCP Catalog page lists 207 servers at the time represented by the source content
Obot 替代品
Docker MCP Enterprise Gateway
www.docker.com
Docker MCP Enterprise Gateway governs how MCP-compatible clients access approved servers and tools. It applies identity-aware policy, supplies credentials from an approved secret store, and records tool-call decisions for enterprise teams.
treg.to
treg.to
treg.to is an API gateway and MCP server for calling tools and data providers through one endpoint and credential. It helps agents and developers use cataloged SEO, enrichment, advertising, social, scraping, and other data tools without managing separate provider keys for every metered call.
Sequel
sequel.sh
Sequel is an AI data analyst and secure data layer for agents. It connects databases, warehouses, analytics platforms, spreadsheets, and SaaS tools so Claude, Cursor, ChatGPT, and other MCP-capable agents can answer questions using shared data connections and definitions.
Workato
workato.com
面向企业级智能体 AI 的集成与编排平台
Edgee
www.edgee.ai
Edgee is an agent gateway for coding teams that reduces token usage, routes requests across models, and provides usage visibility. Its Compression V2 works as a drop-in CLI layer for Claude Code, Codex, OpenCode, Cursor, and other supported agents.
OneCLI
onecli.sh
OneCLI is an open-source platform for giving employees personal AI agents that can work across company tools. It provides isolated sandboxes, gateway-enforced policies, scoped credentials, approvals, and audit logs for team use.