kong
by
Kong

Description: 🦍 The API and AI Gateway

View on GitHub ↗

Summary Information

Updated 17 minutes ago
Added to GitGenius on February 14th, 2025
Created on November 17th, 2014
Open Issues & Pull Requests: 155 (+0)
Number of forks: 5,171
Total Stargazers: 43,763 (+0)
Total Subscribers: 995 (-1)

Issue Activity (beta)

Open issues: 64
New in 7 days: 0
Closed in 7 days: 1
Avg open age: 384 days
Stale 30+ days: 59
Stale 90+ days: 54

Recent activity

Opened in 7 days: 0
Closed in 7 days: 0
Comments in 7 days: 0
Events in 7 days: 0

Top labels

  • pending author feedback (201)
  • stale (152)
  • area/ingress-controller (28)
  • area/kubernetes (26)
  • bug (20)
  • stale/pending revisit (20)
  • core/balancer (12)
  • core/proxy (12)

Repository Insights (GitGenius)

Median issue/PR response: 2.0 days
Mean response time: 197.8 days
90th percentile: 730.9 days
Tracked items: 398

Most active contributors

Detailed Description

Kong is a cloud-native API and AI Gateway written in Lua that serves as a central orchestration layer for microservices, conventional APIs, and agentic LLM and MCP traffic. The project is distinguished by its high performance, extensibility through plugins, and comprehensive feature set for routing, load balancing, authentication, and traffic management across distributed systems.

The gateway provides core functionality including advanced routing and load balancing configurable via RESTful admin API or declarative configuration, authentication and authorization through JWT, basic auth, OAuth, and ACLs, and proxy capabilities with SSL/TLS termination for both L4 and L7 traffic. Kong's plugin architecture enables enforcement of traffic controls, rate limiting, request and response transformations, logging, and monitoring. The project includes over 60 AI-specific features such as AI observability, semantic security and caching, and semantic routing. A universal LLM API routes traffic across multiple providers including OpenAI, Anthropic, GCP Gemini, AWS Bedrock, Azure AI, Databricks, Mistral, and Huggingface. MCP traffic governance, security, and observability capabilities are also built in, along with MCP autogeneration from RESTful APIs.

Kong runs natively on Kubernetes through its official Kubernetes Ingress Controller and supports sophisticated deployment models including Declarative Databaseless Deployment and Hybrid Deployment with control plane and data plane separation. The project maintains a Plugin Hub showcasing Kong Inc. and community-developed plugins like AWS Lambda, Correlation ID, and Response Transformer. Developers can build plugins in Lua, Go, or JavaScript using the Plugin Development Kit.

Activity tracking shows the repository maintains a median issue and pull request response latency of 48.2 hours across 398 tracked items, though mean latency extends to 4746.4 hours indicating some older items. The most active issue labels are pending author feedback with 168 occurrences and stale with 130 occurrences, with area/ingress-controller tracking 23 active items. Top contributors chronolaw, chobits, and StarlightIbuki have driven 218, 149, and 121 tracked events respectively. The repository shares overlapping contributors with kubernetes/minikube, thanos-io/thanos, and prometheus/prometheus, indicating cross-project collaboration within the cloud-native ecosystem.

Kong offers both open-source community editions and commercial subscriptions through Kong Konnect Cloud, which provides managed SaaS control plane functionality, real-time analytics, service catalogs, and developer portals. The project follows semantic versioning and maintains comprehensive documentation, a community forum, Slack channel, and blog for user engagement and knowledge sharing.

kong
by
KongKong/kong

Repository Details

Fetching additional details & charts...