생산 준비가 된 AI 시스템을 어떻게 구축합니까?
hackernews
|
|
🔬 연구
#ai 시스템
#ai 에이전트
#review
#struere
#typescript
#프로덕션 준비
원문 출처: hackernews · Genesis Park에서 요약 및 분석
요약
Struere는 내장형 데이터 레이어와 동적 시스템 프롬프트를 기반으로, TypeScript로 에이전트를 정의하고 HTTP API를 통해 제어할 수 있는 AI 에이전트 플랫폼입니다. 이 플랫폼은 데이터에 인식이 있는 CRUD 연산, 역할 기반 권한 제어, 웹훅을 통한 외부 연동 등을 지원하며, 생산 수준의 환경에서 에이전트를 구축하기 위한 포괄적인 가이드를 제공합니다. 또한 WhatsApp, Google 캘린더, Airtable 등과의 통합을 통해 다양한 외부 서비스와 연동된 자동화 워크플로우를 구현할 수 있습니다.
본문
# Struere > AI agent platform with a built-in data layer, dynamic system prompts, event-driven automation, and integrations. Define agents as TypeScript code, talk to them via HTTP API. ## How to read these docs This is the documentation index. Each URL below returns plain-text markdown. Fetch any URL directly to read that page. URLs are exact — do not modify them. - To read all docs in one file: https://docs.struere.dev/llms-full.txt - To read a topic: fetch a section file below - To read a specific page: fetch its URL from the page list below - OpenAPI spec: https://docs.struere.dev/openapi.yaml ## Section files https://docs.struere.dev/llms-api.txt — Chat API, HTTP endpoints, webhooks https://docs.struere.dev/llms-sdk.txt — Agent, data type, role, automation definitions https://docs.struere.dev/llms-tools.txt — Built-in tools, custom tools, system prompt templates https://docs.struere.dev/llms-platform.txt — Data, agents, automations, events, permissions, evals https://docs.struere.dev/llms-integrations.txt — WhatsApp, Google Calendar, Flow payments, Airtable https://docs.struere.dev/llms-cli.txt — CLI reference https://docs.struere.dev/skill — Developer skill: behavioral rules, gotchas, decision frameworks ## All pages Section headers below are labels only — they are NOT part of the URL path. ### Getting Started https://docs.struere.dev/introduction.md — Introduction: What is Struere and why use it https://docs.struere.dev/getting-started.md — Getting Started: Install Struere and create your first agent https://docs.struere.dev/why-struere.md — Why Struere: When to use Struere and how it compares to building agents from scratch ### API Reference https://docs.struere.dev/api/overview.md — API Overview: HTTP endpoints for interacting with Struere https://docs.struere.dev/api/chat.md — Chat API: Send messages to agents via HTTP https://docs.struere.dev/api/data.md — Data API: CRUD operations for entities via HTTP https://docs.struere.dev/api/webhooks.md — Webhooks: Inbound webhook endpoints for external integrations ### SDK https://docs.struere.dev/sdk/overview.md — SDK Overview: TypeScript SDK for defining agents, data, roles, and automations https://docs.struere.dev/sdk/define-data.md — defineData: Define data type schemas for your domain https://docs.struere.dev/sdk/define-role.md — defineRole: Create roles with policies, scope rules, and field masks https://docs.struere.dev/sdk/define-agent.md — defineAgent: Create and configure AI agent definitions https://docs.struere.dev/sdk/define-trigger.md — defineTrigger: Define event-driven automation rules https://docs.struere.dev/sdk/define-tools.md — defineTools: Create custom tool handlers for agents ### Tools https://docs.struere.dev/tools/built-in-tools.md — Built-in Tools: Pre-built tools available to all agents https://docs.struere.dev/tools/custom-tools.md — Custom Tools: Build custom tool handlers executed on the tool executor service https://docs.struere.dev/tools/system-prompt-templates.md — System Prompt Templates: Dynamic variables and embedded queries in system prompts ### Platform Concepts https://docs.struere.dev/platform/data.md — Data: Domain data with permission-aware CRUD operations https://docs.struere.dev/platform/agents.md — Agents: AI agent configuration and execution https://docs.struere.dev/platform/triggers.md — Automations: Automated workflows triggered by data changes https://docs.struere.dev/platform/events.md — Events: Audit logging and event-driven architecture https://docs.struere.dev/platform/permissions.md — Permissions: Role-based access control with row and column security https://docs.struere.dev/platform/environment-isolation.md — Environment Isolation: Development, production, and eval data separation https://docs.struere.dev/platform/evals.md — Evaluations: Test agent behavior with automated assertions and LLM-as-judge scoring https://docs.struere.dev/platform/studio.md — Studio: Browser-based AI coding environment with multi-provider model selection and custom API key support https://docs.struere.dev/platform/dashboard.md — Dashboard: Navigate and use the Struere dashboard to manage your agents and data https://docs.struere.dev/platform/billing.md — Billing and Credits: How credit-based billing works, pricing, and managing costs https://docs.struere.dev/platform/users.md — User Management: Managing users, roles, and organization membership ### Integrations https://docs.struere.dev/integrations/whatsapp.md — WhatsApp Integration: WhatsApp messaging integration via Kapso https://docs.struere.dev/integrations/flow-payments.md — Flow Payments: Payment processing with Flow.cl integration https://docs.struere.dev/integrations/google-calendar.md — Google Calendar: Connect Google Calendar for scheduling, availability checks, and event management https://docs.struere.dev/integrations/embeddable-widget.md — Embeddable Chat Widget: Add a Struere AI chatbot to any website with a single script tag https://docs.struere.dev/integrati
Genesis Park 편집팀이 AI를 활용하여 작성한 분석입니다. 원문은 출처 링크를 통해 확인할 수 있습니다.
공유