Control AI access before it reaches your systems.
Eve is Eden's AI gateway. It routes requests to models, tools, APIs, and databases while enforcing permissions, policy, PII controls, budgets, and telemetry in one path.
AI apps and agents are getting direct paths into production.
Every direct path creates another place to manage credentials, permissions, spend, sensitive data, observability, and audit trails. Eve gives those requests one governed path instead.
Access spreads
Keys, users, apps, agents, and tools all need access to the same systems.
Risk hides in requests
Sensitive data, dangerous commands, and policy violations often appear after the request is already moving.
Cost loses context
Token spend and infrastructure load are hard to explain without tying each request to a model, app, user, or agent.
One governed path between AI and production
Apps and agents connect to Eve. Eve evaluates the request, applies policy, and routes it to the model, database, API, or tool it needs.
Governed request path
Eve evaluates identity, policy, sensitive data, budgets, routing, and telemetry before any backend sees the request.
Connect the systems your AI needs to use.
Eve uses Eden's endpoint marketplace to connect models, databases, APIs, agents, and infrastructure through one governed access path.
Browse and connect databases, LLMs, agents, APIs, and infrastructure.
Amazon RDS
AWS managed relational database service supporting PostgreSQL, MySQL, MariaDB, Oracle, and SQL Server with automated backups and scaling.
Amazon S3
Amazon Simple Storage Service for scalable object storage. Manage buckets, upload and download objects, and configure access policies.
Anthropic
Claude models for safe, helpful AI assistants with large context windows.
AWS
Connect to Amazon Web Services. Manage resources across S3, EC2, RDS, and other AWS services via CLI-compatible configuration.
AWS Lambda
Run serverless functions on AWS Lambda. Invoke any Lambda function with full request/response support.
Azure
Connect to Microsoft Azure cloud services. Manage resources across Blob Storage, Cosmos DB, Azure SQL, and other Azure services.
Azure OpenAI (classic deployments)
Microsoft Azure OpenAI service via the per-deployment URL ({endpoint}/openai/deployments/{deployment}/chat/completions?api-version=…). Auth via api-key header; deployment in the path selects the model.
Cassandra
Distributed NoSQL database for handling large volumes of data across commodity servers.
Claude Code
Anthropic's agentic coding tool for terminal-based software work.
ClickHouse
Column-oriented OLAP database for real-time analytics on large datasets.
Cline
Autonomous coding agent that can use custom model providers.
Codex
OpenAI coding agent workflow for software engineering tasks.
Cursor
AI code editor with configurable model provider settings.
Databricks
Unified analytics platform for data engineering, data science, and machine learning on a lakehouse architecture.
Datadog
Cloud monitoring and observability platform. Query metrics, logs, events, monitors, and dashboards.
Descript
AI video and podcast editing workflow with agent-style automation surfaces.
ElastiCache
AWS managed Redis-compatible in-memory caching service with automatic failover and scaling.
Eraser
AI-powered diagram generation. Create entity relationship diagrams, flowcharts, sequence diagrams, cloud architecture diagrams, and BPMN diagrams from natural language or DSL.
GitHub
Connect to GitHub repositories. Browse code, manage issues, pull requests, and automate workflows through the GitHub API.
GitLab
Connect to GitLab projects. Browse repositories, manage issues and merge requests, and work with pipelines through the GitLab API.
Google Workspace
Connect Gmail, Drive, Calendar, Docs, Sheets, Chat, and more through one OAuth-backed Google Workspace endpoint.
GraphQL
Connect to GraphQL APIs with schema introspection and query building.
Hermes Agent
Open-source, self-improving local agent from Nous Research.
Kilo Code
Open-source coding agent that works across editor and terminal workflows.
MongoDB
Document-oriented NoSQL database for modern applications with flexible schemas.
MySQL
Popular open-source relational database trusted by millions of applications worldwide.
Ollama
Run open-source LLMs locally. Supports Llama, Mistral, CodeLlama, and more.
Open WebUI
Self-hosted LLM frontend and agent platform with OpenAI-compatible providers.
OpenAI
GPT-4, GPT-3.5, and other models from OpenAI for chat, completion, and embeddings.
OpenClaw
Open-source AI agent that connects to messaging and workflow surfaces.
OpenRouter
Unified API gateway for accessing hundreds of AI models from multiple providers.
Oracle
Enterprise-grade relational database with comprehensive features for mission-critical workloads.
Pieces
Private, on-device memory layer for developer workflows.
Pinecone
Vector database for building high-performance similarity search and AI applications.
PostgreSQL
The world's most advanced open-source relational database. Full SQL support with ACID compliance.
PostHog
Connect to PostHog product analytics. Query events with HogQL, retrieve insights, analyze funnels and trends, and manage feature flags.
Redis
In-memory data store for caching, real-time analytics, and message brokering.
REST API
Connect to any REST API with configurable authentication and request mapping.
Roo Code
Autonomous coding agent for editor-based development workflows.
Salesforce
Connect to Salesforce CRM. Query and manage objects, run SOQL queries, and sync data across your Salesforce organization.
Snowflake
Cloud data platform for data warehousing, data lakes, and data sharing with near-unlimited scale.
SQL Server
Microsoft's enterprise relational database with built-in analytics and reporting.
Tavily
AI-powered web search and research API. Enable real-time web search and deep research capabilities for your AI agents.
Weaviate
Open-source vector database for AI-powered search, retrieval-augmented generation, and semantic applications.
Search and scroll the marketplace preview here. The live dashboard includes setup flows for each endpoint.
What Eve does in the request path
Authenticate, route, enforce, and analyze each request before it reaches the systems behind your AI workloads.
Authenticate every caller
Put users, apps, agents, and keys behind one access model before they reach production systems.
- User, robot, app, and agent identity
- Organization-scoped credentials
- Resource-level permissions
Route every request
Send traffic to the right model, API, tool, or database without hardwiring every client to one backend.
- OpenAI, Anthropic, Ollama, and OpenRouter
- Native database protocols
- APIs and functions as governed endpoints
Enforce policy in the path
Block, warn, mask, or observe risky requests before they execute.
- Command policy by endpoint
- PII detection and redaction
- Audit trails for regulated workloads
Analyze and optimize usage
Attach policy, routing, cost, latency, and consumer context to the same request record so teams can control spend and performance.
- Metrics, traces, and logs from the gateway
- Budgets by app, user, agent, and model
- Spend tied to each request
How It Works
Point AI and data traffic through Eve
Connect apps, agents, and internal services to Eve instead of directly to every backend.
Eve evaluates each request
Auth, policy, PII controls, budgets, routing, and telemetry run in the gateway path.
Every action gets context
Requests are tied back to the app, user, agent, key, endpoint, model, and policy decision.
Performance & Infrastructure
Built for production workloads at scale.
Target added latency per operation at the wire protocol level
Thread-sharded atomics with cache-line padding for low-overhead telemetry
Built on Cloudflare's Pingora with round-robin, HTTP/2, and health checking
Block, warn, or observe — command-level control per endpoint with 15+ presets
Databases, LLMs, APIs, Lambda functions, vector stores, tools endpoints
DogStatsD (UDP) and Prometheus exposition — plug into your existing stack
Give AI one governed path.
Connect your first system and see Eve govern access in real time.