skilder
Verified Safeby skilder-ai
Overview
Skilder is an infrastructure layer for AI agent tooling, providing a private tool registry with embedded runtimes for connecting AI agents (e.g., LangChain, CrewAI) to various services (MCP, REST APIs, custom functions) across distributed environments.
Installation
docker compose up -d --pull=alwaysEnvironment Variables
- DGRAPH_URL
- NATS_SERVERS
- CORS_ORIGINS
- JWT_PRIVATE_KEY_PATH
- JWT_PUBLIC_KEY_PATH
- SYSTEM_KEY
- ENCRYPTION_KEY
- LOG_LEVEL
- EXPOSED_REMOTE_MCP
- EXPOSED_NATS_SERVERS
- RUNTIME_NAME
- REMOTE_PORT
- ROOTS
- FRONTEND_URL
- PASSWORD_MIN_LENGTH
- JWT_ISSUER
- JWT_ACCESS_TOKEN_TTL
- JWT_REFRESH_TOKEN_TTL
- RATE_LIMIT_MAX
- RATE_LIMIT_WINDOW
Security Notes
The server demonstrates strong security practices including JWT-based authentication with database validation against stale tokens, periodic re-validation of subscription access, robust cryptographic key generation with strict file permissions, encrypted OAuth state with nonce-checking to prevent CSRF and replay attacks, and rate limiting on authentication and OAuth initiation endpoints. Dgraph Alpha's development configuration allows all IPs, but production uses a restricted whitelist. The Python client examples have hardcoded local paths for the runtime which is a configuration oversight for general use, but not a core security vulnerability in the server itself. Content Security Policy and HSTS are disabled in development, as expected, but should be enabled for production.
Similar Servers
MCPJungle
A self-hosted gateway and registry for Model Context Protocol (MCP) servers, allowing AI agents to discover and consume tools from a central location.
toolsdk-mcp-registry
An open, structured registry and gateway for discovering, securing, and executing Model Context Protocol (MCP) servers and their tools.
agentor
Deploy a scalable AI Agent server with tool integration and async streaming capabilities using LitServe, compatible with the Celesto AI platform.
Polymcp
A comprehensive toolkit and agent framework for building Model Context Protocol (MCP) servers and orchestrating them with Large Language Models (LLMs) across Python and TypeScript environments.