Agentic Fabriq

Agentic Fabriq is a Y Combinator-backed security platform focused on identity, governance, and access control for AI agents. The platform addresses the challenge that existing identity infrastructure was not designed for autonomous, non-human actors operating within enterprise and developer environments. The platform serves as a centralized hub through which AI agents route their connections to external tools and services, rather than connecting directly. Fabriq handles authentication, credential management, and policy enforcement on behalf of those agents. Core capabilities include: - Agent Identity Management: Creation and lifecycle management of unique identities for each AI agent, with support for delegated user identity and permission constraints. - Fine-Grained Authorization: Resource-level permission policies that govern what each agent can access and do, evaluated per agent and per user in real time. - Credential and Token Management: A secure vault that prevents agents from storing sensitive secrets directly; Fabriq handles token exchange and integrates with existing identity providers. - Audit Logging and Monitoring: Immutable, real-time logs of every agent action, enabling compliance reporting and incident investigation. Agentic Fabriq supports two primary deployment contexts: internal enterprise agents acting on behalf of employees, and developer-facing deployments where agents act on behalf of external customers without the product holding user tokens. The platform integrates with major agent frameworks including LangChain, CrewAI, AutoGen, and OpenAI Agents, and connects to tools such as Gmail, Slack, GitHub, Google Drive, Microsoft 365, and others via SDKs and REST APIs. The founding team includes alumni from MIT.