EnterpriseAgentFramework
ReachAI企业级智能体开发平台：快速、安全完成已有业务系统智能化改造，让 AI 在 OA、ERP、CRM 等原系统中查数据、填表单、办业务。ReachAI: Quickly and securely bring A
What is EnterpriseAgentFramework?
ReachAI企业级智能体开发平台：快速、安全完成已有业务系统智能化改造，让 AI 在 OA、ERP、CRM 等原系统中查数据、填表单、办业务。ReachAI: Quickly and securely bring AI to existing enterprise systems, enabling AI to query data, fill out forms, and execute business tasks directly within OA, ERP, CRM, and other business applications.
Overview of EnterpriseAgentFramework
What it is
EnterpriseAgentFramework (ReachAI) is an open-source enterprise agent development platform designed to quickly and securely bring AI to existing enterprise systems. It enables AI to query data, fill out forms, and execute business tasks directly within OA, ERP, CRM, eHR, procurement, work orders, contracts, and other business applications without requiring system migration. The platform covers the full lifecycle from system connection, scanning, AI-assisted coding, capability publishing, business acceptance, to production governance.
Key features
- Scans existing Java enterprise systems to identify pages, routes, components, APIs, and business capabilities, producing an evidence-based system capability map.
- Recommends up to three candidate transformation items per selected page, with code references, implementation references, and acceptance criteria.
- Converts transformation requirements into standardized engineering tasks for AI Coding tools (Codex, Trae, Cursor, Claude Code) and verifies code, service, SDK callbacks, and real business results.
- Provides a Java SDK (JDK 8 compatible) and Spring Boot 2 Starter with
@ReachCapabilityannotations to register business methods as governable capabilities. - Combines Agent (for flexible intent understanding and task planning) with Workflow (for stable, pre-verified execution of mature business processes).
- Embeds AI assistant into original business pages via a declarative protocol (Page Bridge), avoiding screenshot recognition or mouse simulation.
- Offers an Agent Operations Center that aggregates project-level metrics: agents, run samples, identifiable users, tokens, technical completion rate, risk items, and business system contributions.
Use cases
- OA: “Summarize my pending tasks and open the most urgent one.” - eHR: “Query this month’s abnormal attendance and filter out unprocessed records.” - Procurement: “Create an office supplies purchase request and let me confirm before submission.” - CRM: “Generate a follow-up plan based on this week’s customer activity.”
Getting started
See the repository README for installation and quick-start instructions.
License and pricing
Licensed under MIT. The project is open source.
Based on the project's public README and repository metadata.
