Problem

Guild addresses the operational and security breakdown created when AI agents multiply across an enterprise without centralized governance. Agents running on separate servers or laptops can create shadow AI, security blind spots, fragmented accountability, and runaway inference costs: Guild reports an early-customer case in which one engineer exhausted an entire month’s budget in 12 hours. The central pain is not merely building an agent, but knowing which agent accessed which system, what it did, and how much it cost.

The clearest killer use case is governing autonomous engineering workflows such as pull-request triage, dependency upgrades, and deployment approvals. Without shared identity, least-privilege access, and an audit trail, a failed agent action or infinite build loop can leave platform teams reconstructing events from raw logs. Guild’s value is to make these workflows safe enough to run continuously in production rather than treating each agent as an isolated script.

Product / Service

Guild is a neutral control plane for AI agents: a centralized infrastructure layer through which agents execute, access tools, and are monitored. Each execution runs through Guild, where the agent’s identity is enforced, permissions and OAuth credentials are scoped, and prompts, model calls, tool invocations, and system interactions are recorded. The platform provides cost dashboards, mediated LLM access through a proxy, policy and approval controls, and integrations with systems such as GitHub, Jira, Slack, and Google services.

The product is designed to work across models and frameworks, including Anthropic, OpenAI, Google, open-source models, LangChain, CrewAI, custom Python, and Guild’s own TypeScript SDK. It also includes a governed agent catalog where teams can publish, discover, install, fork, and customize agents inside workspace-scoped environments. The delivery model is commercial software with a free entry point, a stated $199-per-month Starter tier, and enterprise features such as SSO, SCIM, RBAC, audit logs, retention, and SLAs.

Market

Guild competes in the emerging enterprise AI-agent infrastructure, governance, security, and observability market, positioning itself specifically as an agent control plane rather than as another model provider or agent-building framework. Its direct category is still developing, while adjacent alternatives include Microsoft Foundry Agent Service, a managed platform for building, deploying, and scaling agents, and LangSmith, which focuses on observability for AI workflows. Guild’s stated differentiation is inline runtime governance, vendor and framework neutrality, centralized cost control, and a shared organizational layer for every agent.

The company has meaningful early traction but no disclosed revenue or customer-count figure in the available evidence. Guild announced $44 million in total funding, comprising a $30 million latest round and $14 million seed round led by GV, and Axios reported a $300 million valuation. Its general availability followed in April 2026, with paid pricing already published, while company-reported early deployments include automated pull-request risk analysis, a 200-plus-agent global support fleet, 72% automation of routine dependency upgrades, and more than $18,000 in monthly savings in one example. These signals indicate an early commercial launch with customer validation, but not yet a publicly quantified revenue scale.