iaurora 1.0.24 → 1.0.26

package/.env.example

@@ -1,29 +1,454 @@
-# Aurora Gateway — Environment Variables
-# Copy or rename this file to .env and fill in your values.
-# The .env file must be in the same directory as your config YAML.
-
-# REQUIRED: Gateway security key (set before exposing outside localhost)
-AURORA_MASTER_KEY=replace-with-secure-key
-
-# REQUIRED: At least one provider API key (uncomment one or more)
-# GROQ_API_KEY=gsk_...
-# OPENAI_API_KEY=sk-...
-# ANTHROPIC_API_KEY=sk-ant-...
-# GEMINI_API_KEY=...
-# DEEPSEEK_API_KEY=...
-# OPENROUTER_API_KEY=sk-or-...
-# XAI_API_KEY=...
-# AZURE_API_KEY=...
-# AZURE_BASE_URL=https://your-resource.openai.azure.com/...
-# OLLAMA_BASE_URL=http://localhost:11434/v1
-# VLLM_BASE_URL=http://localhost:8000/v1
-
-# OPTIONAL: Server settings
-# PORT=8080
-# BASE_PATH=/
-# STORAGE_TYPE=sqlite
-# METRICS_ENABLED=false
-
-# OPTIONAL: Redis (for response cache)
-# REDIS_URL=redis://localhost:6379
-# RESPONSE_CACHE_SIMPLE_ENABLED=true
+# Aurora Gateway Environment Template
+#
+# Copy this file to .env in this directory for local Docker Compose runs.
+# Keep real secrets in .env only; this template must stay secret-free.
+#
+# Syntax notes:
+# - Values are examples/defaults unless marked required for a feature.
+# - Durations accept seconds for many numeric settings, or Go durations where noted.
+# - Provider instances use <PROVIDER>_<SUFFIX>_*; suffix underscores become hyphens.
+# - Preferred setup is YAML for structure and env vars for secrets/runtime values.
+#   Use AURORA_CONFIG_PATH=configs/editions/oss.example.yaml for minimal local,
+#   configs/editions/oss.local-power.example.yaml for Redis-backed local power use,
+#   or configs/editions/oss.team.example.yaml for Postgres/Redis/Qdrant team deployments.
+# - This template documents OSS runtime settings only.
+
+# =============================================================================
+# Server & Routing
+# =============================================================================
+
+# HTTP port for the gateway (default: 8080)
+# PORT=8080
+
+# Mount the gateway under a path prefix, e.g. https://example.com/g/ (default: /)
+# BASE_PATH=/g
+
+# Maximum request body size. Accepts "10M", "1G", "500K" (default: 10M)
+# BODY_SIZE_LIMIT=10M
+
+# Enable Swagger UI at /swagger/index.html (default: false in code defaults)
+# SWAGGER_ENABLED=true
+
+# Enable pprof profiling routes at /debug/pprof/* (default: false)
+# PPROF_ENABLED=false
+
+# Enable provider-native passthrough routes under /p/{provider}/{endpoint} (default: true)
+# ENABLE_PASSTHROUGH_ROUTES=true
+
+# Allow optional /p/{provider}/v1/... aliases while keeping /p/{provider}/... canonical (default: true)
+# ALLOW_PASSTHROUGH_V1_ALIAS=true
+
+# Provider types enabled for passthrough routes
+# ENABLED_PASSTHROUGH_PROVIDERS=openai,anthropic,openrouter,zai,vllm
+
+# Expose /v1/messages for native Anthropic-format clients (default: false)
+# ENABLE_ANTHROPIC_INGRESS=false
+
+# =============================================================================
+# Security
+# =============================================================================
+
+# CRITICAL: Set this to secure the gateway from unauthorized access.
+# If unset, the server may run in unsafe mode with a warning.
+# AURORA_MASTER_KEY=replace-with-secure-random-key
+
+# =============================================================================
+# Logging & Runtime Metadata
+# =============================================================================
+
+# Log output format: auto-detect when unset, or set to json/text
+# LOG_FORMAT=text
+
+# Log verbosity: debug, info, warn, or error (default: info)
+# LOG_LEVEL=info
+
+# Include source file/line in logs when supported (default: false)
+# LOG_ADD_SOURCE=false
+
+# Optional service metadata added to structured logs
+# SERVICE_NAME=aurora-gateway
+# ENV=development
+
+# =============================================================================
+# HTTP Client Timeouts
+# =============================================================================
+
+# Overall upstream request timeout (default: 600 seconds)
+# HTTP_TIMEOUT=600
+
+# Time to wait for upstream response headers (default: 600 seconds)
+# HTTP_RESPONSE_HEADER_TIMEOUT=600
+
+# =============================================================================
+# Metrics
+# =============================================================================
+
+# Enable Prometheus metrics collection and /metrics endpoint (default: false)
+# METRICS_ENABLED=false
+
+# Metrics endpoint path (default: /metrics)
+# METRICS_ENDPOINT=/metrics
+
+# =============================================================================
+# Admin API & Dashboard
+# =============================================================================
+
+# Enable /admin/api/v1/* REST endpoints (default: true)
+# ADMIN_ENDPOINTS_ENABLED=true
+
+# Enable /admin/dashboard UI (default: true; requires ADMIN_ENDPOINTS_ENABLED=true)
+# ADMIN_UI_ENABLED=true
+
+# Dashboard implementation variant (default: react)
+
+
+# =============================================================================
+# Config Files & Dashboard Overrides
+# =============================================================================
+
+# Optional config path override. When unset, the gateway loads configs/config.yaml.
+# AURORA_CONFIG_PATH=configs/editions/oss.example.yaml
+
+# Local path to manual fallback rules for direct/air runs. Docker Compose overrides this to /app/configs/fallback.local.json.
+# FALLBACK_MANUAL_RULES_PATH=./.aurora.local/fallback.local.json
+
+# Default translated-route fallback mode: auto, manual, or off (default: manual)
+# FEATURE_FALLBACK_MODE=manual
+
+# How often to refresh persisted workflows from storage (default: 1m)
+# WORKFLOW_REFRESH_INTERVAL=1m
+
+# =============================================================================
+# Model Registry & Model Access
+# =============================================================================
+
+# External model metadata registry. Leave empty to disable upstream fetch.
+# MODEL_LIST_URL=https://auroraix.github.io/aurora/models.json
+
+# Local model registry snapshot and user pricing overrides
+# MODEL_LIST_LOCAL_PATH=data/models.local.json
+# MODEL_LIST_USER_OVERRIDES_PATH=data/user_pricing.yaml
+
+# Process-wide default for provider models when no persisted override exists (default: true)
+# MODELS_ENABLED_BY_DEFAULT=true
+
+# Enable persisted model overrides and dashboard editing (default: true)
+# MODEL_OVERRIDES_ENABLED=true
+
+# Hide provider models from GET /v1/models and expose only enabled aliases (default: false)
+# KEEP_ONLY_ALIASES_AT_MODELS_ENDPOINT=false
+
+# How configured provider model lists affect inventory: fallback or allowlist (default: fallback)
+# CONFIGURED_PROVIDER_MODELS_MODE=fallback
+
+# =============================================================================
+# Model Cache
+# =============================================================================
+
+# Dashboard/runtime cache toggle used by admin settings views.
+# CACHE_ENABLED=true
+
+# Local filesystem cache directory for model metadata (default: .cache when enabled)
+# AURORA_CACHE_DIR=.cache
+
+# Redis-backed model cache. Set REDIS_URL to use Redis instead of local cache.
+# For Docker Compose app profile use redis://redis:6379; for host-only runs use redis://localhost:6379.
+# REDIS_URL=redis://redis:6379
+# REDIS_KEY_MODELS=aurora:models
+# REDIS_TTL_MODELS=86400
+
+# How often to refresh model registry cache in seconds (default: 3600)
+# CACHE_REFRESH_INTERVAL=3600
+
+# Optional Redis connection helper values used by local/docker tooling.
+# REDIS_CONNECTION_LIMIT=20
+# REDIS_PORT=6379
+
+# =============================================================================
+# Response Cache - Simple Redis Cache
+# =============================================================================
+
+# Opt-in when config.yaml has no cache.response.simple block (env-only deploys)
+# RESPONSE_CACHE_SIMPLE_ENABLED=true
+# REDIS_KEY_RESPONSES=aurora:response:
+# REDIS_TTL_RESPONSES=3600
+
+# =============================================================================
+# Response Cache - Semantic Cache
+# =============================================================================
+
+# Opt-in when config.yaml has no cache.response.semantic block (env-only deploys)
+# SEMANTIC_CACHE_ENABLED=true
+
+# Similarity thresholds between 0 and 1
+# SEMANTIC_CACHE_THRESHOLD=0.92
+# SEMANTIC_CACHE_PROMPT_SIMILARITY=0.90
+
+# Semantic cache entry TTL in seconds and recent conversation messages to embed
+# SEMANTIC_CACHE_TTL=3600
+# SEMANTIC_CACHE_MAX_CONV_MESSAGES=3
+
+# Exclude the system prompt from semantic cache keys (default: false)
+# SEMANTIC_CACHE_EXCLUDE_SYSTEM_PROMPT=false
+
+# Embedder used by semantic cache
+# SEMANTIC_CACHE_EMBEDDER_PROVIDER=openai
+# SEMANTIC_CACHE_EMBEDDER_MODEL=text-embedding-3-small
+
+# Vector store backend: qdrant, pgvector, pinecone, or weaviate
+# SEMANTIC_CACHE_VECTOR_STORE_TYPE=qdrant
+
+# Qdrant semantic cache backend
+# SEMANTIC_CACHE_QDRANT_URL=http://localhost:6333
+# SEMANTIC_CACHE_QDRANT_COLLECTION=aurora_semantic
+# SEMANTIC_CACHE_QDRANT_API_KEY=
+
+# pgvector semantic cache backend
+# SEMANTIC_CACHE_PGVECTOR_URL=postgres://user:pass@localhost:5432/aurora
+# SEMANTIC_CACHE_PGVECTOR_TABLE=aurora_semantic_cache
+# SEMANTIC_CACHE_PGVECTOR_DIMENSION=1536
+
+# Pinecone semantic cache backend
+# SEMANTIC_CACHE_PINECONE_HOST=https://your-index.svc.region.pinecone.io
+# SEMANTIC_CACHE_PINECONE_API_KEY=
+# SEMANTIC_CACHE_PINECONE_NAMESPACE=
+# SEMANTIC_CACHE_PINECONE_DIMENSION=1536
+
+# Weaviate semantic cache backend
+# SEMANTIC_CACHE_WEAVIATE_URL=http://localhost:8080
+# SEMANTIC_CACHE_WEAVIATE_CLASS=AuroraSemanticCache
+# SEMANTIC_CACHE_WEAVIATE_API_KEY=
+
+# =============================================================================
+# Storage
+# =============================================================================
+
+# Storage type: sqlite (default), postgresql, or mongodb
+# STORAGE_TYPE=sqlite
+
+# SQLite storage
+# SQLITE_PATH=data/aurora.db
+
+# PostgreSQL storage
+# For Docker Compose app profile use postgres service DNS; for host-only runs use localhost.
+# POSTGRES_URL=postgres://aurora:aurora@postgres:5432/aurora?sslmode=disable
+# POSTGRES_MAX_CONNS=10
+# POSTGRES_DB=aurora
+
+# MongoDB storage
+# MONGODB_URL=mongodb://localhost:27017/aurora
+# MONGODB_DATABASE=aurora
+
+# =============================================================================
+# Audit Logging
+# =============================================================================
+
+# Enable audit logging to configured storage (default: false)
+# LOGGING_ENABLED=false
+
+# WARNING: Bodies may contain PII, prompts, or sensitive data.
+# LOGGING_LOG_BODIES=false
+
+# Sensitive headers are automatically redacted.
+# LOGGING_LOG_HEADERS=false
+
+# Log only model interactions, skipping /health, /metrics, /admin (default: true)
+# LOGGING_ONLY_MODEL_INTERACTIONS=true
+
+# In-memory audit log queue capacity and flush cadence
+# LOGGING_BUFFER_SIZE=1000
+# LOGGING_FLUSH_INTERVAL=5
+
+# Auto-delete logs older than N days, 0 = keep forever (default: 30)
+# LOGGING_RETENTION_DAYS=30
+
+# =============================================================================
+# Usage Tracking
+# =============================================================================
+
+# Enable token usage tracking (default: true)
+# USAGE_ENABLED=true
+
+# Enable admin usage pricing recalculation action (default: true)
+# USAGE_PRICING_RECALCULATION_ENABLED=true
+
+# Add stream_options.include_usage=true to streaming OpenAI-compatible requests (default: true)
+# ENFORCE_RETURNING_USAGE_DATA=true
+
+# In-memory usage queue capacity and flush cadence
+# USAGE_BUFFER_SIZE=1000
+# USAGE_FLUSH_INTERVAL=5
+
+# Auto-delete usage data older than N days, 0 = keep forever (default: 90)
+# USAGE_RETENTION_DAYS=90
+
+# =============================================================================
+# Guardrails
+# =============================================================================
+
+# Enable configured guardrails globally (default: false)
+# GUARDRAILS_ENABLED=false
+
+# Apply guardrails to inline /v1/batches request items (default: false)
+# ENABLE_GUARDRAILS_FOR_BATCH_PROCESSING=false
+
+# Token Saver
+# =============================================================================
+
+# Caveman-style output compression (default: disabled).
+# Injects a 'talk like caveman' system instruction — drops filler, keeps substance.
+# TOKEN_SAVER_ENABLED=false
+# TOKEN_SAVER_ENDPOINTS=chat_completions
+# TOKEN_SAVER_APPLY_STREAMING=true
+
+# Output style
+# TOKEN_SAVER_OUTPUT_ENABLED=false
+# TOKEN_SAVER_OUTPUT_PROFILE=concise
+
+# Optional include/exclude filters
+# TOKEN_SAVER_MODELS_INCLUDE=
+# TOKEN_SAVER_MODELS_EXCLUDE=
+# TOKEN_SAVER_PROVIDERS_INCLUDE=
+# TOKEN_SAVER_PROVIDERS_EXCLUDE=
+
+# Error/header/audit behavior
+# TOKEN_SAVER_ON_ERROR=allow
+# TOKEN_SAVER_EMIT_HEADERS=true
+# TOKEN_SAVER_AUDIT_ENABLED=true
+
+# =============================================================================
+# Resilience
+# =============================================================================
+
+# Retry attempts for upstream provider calls (default: 3)
+# RETRY_MAX_RETRIES=3
+# RETRY_INITIAL_BACKOFF=1s
+# RETRY_MAX_BACKOFF=30s
+# RETRY_BACKOFF_FACTOR=2.0
+# RETRY_JITTER_FACTOR=0.1
+
+# Circuit breaker settings
+# CIRCUIT_BREAKER_FAILURE_THRESHOLD=5
+# CIRCUIT_BREAKER_SUCCESS_THRESHOLD=2
+# CIRCUIT_BREAKER_TIMEOUT=30s
+
+# =============================================================================
+# Admin Tools & Combos
+# =============================================================================
+
+# Enable CLI tools integration and whether admin/API can apply tool changes
+# CLI_TOOLS_ENABLED=true
+# CLI_TOOLS_APPLY_ENABLED=false
+
+# Enable combo model calls/workflows (default: true)
+# COMBOS_ENABLED=true
+
+# =============================================================================
+# Provider API Keys & Base URLs
+# =============================================================================
+
+# OpenAI
+# OPENAI_API_KEY=sk-...
+# OPENAI_BASE_URL=https://api.openai.com/v1
+# OPENAI_MODELS=gpt-4o,gpt-4o-mini
+
+# Anthropic
+# ANTHROPIC_API_KEY=sk-ant-...
+# ANTHROPIC_BASE_URL=https://api.anthropic.com/v1
+# ANTHROPIC_DEFAULT_MAX_TOKENS=4096
+# ANTHROPIC_MODELS=claude-sonnet-4-6,claude-opus-4-8
+
+# Google Gemini (OpenAI-compatible endpoint)
+# GEMINI_API_KEY=...
+# GEMINI_BASE_URL=https://generativelanguage.googleapis.com/v1beta/openai
+# GEMINI_MODELS=gemini-2.5-pro,gemini-2.5-flash
+
+# xAI (Grok)
+# XAI_API_KEY=...
+# XAI_BASE_URL=https://api.x.ai/v1
+# XAI_MODELS=grok-3,grok-3-mini
+
+# Groq
+# GROQ_API_KEY=gsk_...
+# GROQ_BASE_URL=https://api.groq.com/openai/v1
+# GROQ_MODELS=llama-3.3-70b-versatile
+
+# OpenRouter
+# OPENROUTER_API_KEY=sk-or-...
+# OPENROUTER_BASE_URL=https://openrouter.ai/api/v1
+# OPENROUTER_MODELS=openai/gpt-oss-120b,anthropic/claude-sonnet-4
+# OPENROUTER_SITE_URL=https://github.com/auroraix/aurora
+# OPENROUTER_APP_NAME=Aurora Gateway
+
+# Z.ai
+# ZAI_API_KEY=...
+# ZAI_BASE_URL=https://api.z.ai/api/paas/v4
+# ZAI_MODELS=glm-4.5
+
+# MiniMax
+# MINIMAX_API_KEY=...
+# MINIMAX_BASE_URL=https://api.minimax.io/v1
+# MINIMAX_MODELS=minimax-m1
+
+# Azure OpenAI
+# AZURE_API_KEY=...
+# AZURE_BASE_URL=https://your-resource.openai.azure.com/openai/deployments/your-deployment
+# AZURE_API_VERSION=2024-10-21
+# AZURE_MODELS=deployment-name
+
+# Oracle
+# ORACLE_API_KEY=...
+# ORACLE_BASE_URL=https://inference.generativeai.us-chicago-1.oci.oraclecloud.com/20231130/actions/v1
+# ORACLE_MODELS=openai.gpt-oss-120b,xai.grok-3
+
+# Ollama (local LLM server; API key optional for secured deployments)
+# OLLAMA_API_KEY=...
+# OLLAMA_BASE_URL=http://localhost:11434/v1
+# OLLAMA_MODELS=llama3.1
+
+# vLLM (OpenAI-compatible server; API key optional)
+# VLLM_API_KEY=<your-vllm-api-key>
+# VLLM_BASE_URL=http://localhost:8000/v1
+# VLLM_MODELS=local-model
+
+# Local/proxy provider examples from project-local .env. Replace with your own values.
+# OPENAI_PROXY_API_KEY=replace-with-openai-proxy-key
+# OPENAI_PROXY_BASE_URL=https://proxy.example.com/openai/v1
+# OPENAI_PROXY_MODELS=gpt-4o-mini
+# GOOGLE_PROXY_API_KEY=replace-with-google-proxy-key
+# GOOGLE_PROXY_BASE_URL=https://proxy.example.com/google/v1beta
+# GOOGLE_PROXY_MODELS=gemini-2.5-flash
+# INCEPTION_API_KEY=replace-with-inception-key
+# INCEPTION_BASE_URL=https://api.inceptionlabs.ai/v1
+# INCEPTION_MODELS=mercury-coder-small
+# THIRDPARTY_API_KEY=replace-with-thirdparty-key
+# THIRDPARTY_BASE_URL=https://example-thirdparty.invalid/v1
+# THIRDPARTY_MODELS=example-model
+
+# Jina embeddings / reranking. Numbered keys can be used for multiple provider instances.
+# JINA_API_KEY=replace-with-jina-key
+# JINA_API_KEY1=replace-with-jina-key-1
+# JINA_API_KEY2=replace-with-jina-key-2
+# JINA_API_KEY3=replace-with-jina-key-3
+# JINA_BASE_URL=https://api.jina.ai/v1
+# JINA_MODELS=jina-embeddings-v3
+
+# =============================================================================
+# Docker Compose / Local Runtime Helpers
+# =============================================================================
+
+# Host/container port mapping helpers used by docker-compose or local scripts
+# AURORA_HOST_PORT=8088
+
+# Host-side paths for mounted config, fallback rules, and CA certificate
+# AURORA_CONFIG_HOST_PATH=.aurora.local/config.yaml
+# AURORA_FALLBACK_HOST_PATH=.aurora.local/fallback.local.json
+# AURORA_CA_PEM_HOST_PATH=.aurora.local/ca.pem
+
+# Local CA path for direct/air runs. Docker Compose overrides this to /app/configs/ca.pem.
+# CA_PEM_PATH=.aurora.local/ca.pem
+
+# General Qdrant settings used by docker-compose/config.yaml expansion.
+# For semantic cache specifically, use SEMANTIC_CACHE_QDRANT_* above.
+# QDRANT_URL=http://qdrant:6333
+# QDRANT_API_KEY=

package/README.md

@@ -1,22 +1,8 @@
 <p align="center">
-  <svg viewBox="0 0 32 32" width="96" height="96" fill="none" xmlns="http://www.w3.org/2000/svg">
-    <path d="M4 20 C8 12, 14 10, 16 16 C18 22, 24 20, 28 12" stroke="#6AD87A" stroke-width="1.8" stroke-linecap="round" fill="none" opacity="0.5" stroke-dasharray="40" stroke-dashoffset="40">
-      <animate attributeName="stroke-dashoffset" from="40" to="0" dur="1.8s" fill="freeze"/>
-    </path>
-    <path d="M2 24 C8 14, 14 8, 16 18 C18 28, 24 22, 30 14" stroke="#6AD87A" stroke-width="1.8" stroke-linecap="round" fill="none" opacity="0.35" stroke-dasharray="50" stroke-dashoffset="50">
-      <animate attributeName="stroke-dashoffset" from="50" to="0" dur="2.2s" begin="0.3s" fill="freeze"/>
-    </path>
-    <path d="M6 16 C10 10, 14 12, 16 14 C18 16, 22 14, 26 10" stroke="#6AD87A" stroke-width="1.8" stroke-linecap="round" fill="none" opacity="0.65" stroke-dasharray="35" stroke-dashoffset="35">
-      <animate attributeName="stroke-dashoffset" from="35" to="0" dur="1.6s" begin="0.6s" fill="freeze"/>
-    </path>
-    <circle cx="16" cy="16" r="2.2" fill="#6AD87A">
-      <animate attributeName="opacity" values="1;0.5;1" dur="2.5s" repeatCount="indefinite"/>
-      <animate attributeName="r" values="2.2;2.8;2.2" dur="2.5s" repeatCount="indefinite"/>
-    </circle>
-  </svg>
+  <img src="https://raw.githubusercontent.com/aurorallm/aurora/main/docs/assets/aurora-logo-animated.svg" width="96" height="96" alt="Aurora Logo">
 </p>
 
-<h1 align="center">Aurora Gateway</h1>
+<h1 align="center">Aurora Gateway OSS — Open-Source AI Gateway | OpenAI &amp; Anthropic Compatible API</h1>
 
 <p align="center">
   <a href="LICENSE"><img src="https://img.shields.io/github/license/aurorallm/aurora" alt="License" height="20"></a>
@@ -26,23 +12,22 @@
   <a href="https://github.com/aurorallm/aurora"><img src="https://img.shields.io/github/stars/aurorallm/aurora" alt="GitHub Stars" height="20"></a>
   <a href="https://discord.gg/YJPrfR9uh"><img src="https://dcbadge.limes.pink/api/server/https://discord.gg/YJPrfR9uh?style=flat" alt="Discord" height="20"></a>
   <img src="https://img.shields.io/docker/pulls/aurorahq/aurora" alt="Docker Pulls" height="20">
-  <a href="https://artifacthub.io/packages/search?repo=aurora"><img src="https://img.shields.io/endpoint?url=https://artifacthub.io/badge/repository/aurora" alt="Artifact Hub" height="20"></a>
+  <a href="https://artifacthub.io/packages/search?repo=aurora-gateway"><img src="https://img.shields.io/endpoint?url=https://artifacthub.io/badge/repository/aurora-gateway" alt="Artifact Hub" height="20"></a>
 </p>
 
-## One gateway. Every provider. Open source.
+<p align="center"><b>Open-source LLM gateway (OSS edition). One API for every AI provider.</b></p>
 
-Route AI traffic through a single OpenAI- and Anthropic-compatible API.
-Self-hosted. No vendor lock-in. Full source visibility.
+<p align="center">Self-hosted. No vendor lock-in. 30+ LLM providers supported.</p>
 
-<a href="docs/assets/dashboard-overview.png">
-  <img src="docs/assets/dashboard-overview.png" alt="Aurora dashboard" width="100%">
+<a href="https://raw.githubusercontent.com/aurorallm/aurora/main/docs/assets/dashboard-overview.png">
+  <img src="https://raw.githubusercontent.com/aurorallm/aurora/main/docs/assets/dashboard-overview.png" alt="Aurora open-source AI gateway admin dashboard showing provider stats and usage metrics" width="100%">
 </a>
 
 ## Quick Start
 
-**Go from zero to running in under a minute.**
+Start routing AI traffic in 60 seconds.
 
-<img src="./install.gif" alt="Install demo" style="max-width:100%;border-radius:12px;margin:12px 0;">
+<img src="https://raw.githubusercontent.com/aurorallm/aurora/main/docs/assets/install.gif" alt="Install demo" style="max-width:100%;border-radius:12px;margin:12px 0;">
 
 **1. Install Aurora Gateway**
 
@@ -82,7 +67,39 @@ curl http://localhost:8080/v1/chat/completions \
 
 The dashboard is at `http://localhost:8080/admin/dashboard`.
 
-**Setup guides:** [npm](https://www.npmjs.com/package/iaurora) · [Docker](https://hub.docker.com/r/aurorahq/aurora) · [Source](https://github.com/aurorallm/aurora)
+**Setup guides:** [npm](https://www.npmjs.com/package/iaurora) · [Docker](https://hub.docker.com/r/aurorahq/aurora) · [Helm/Kubernetes](https://github.com/aurorallm/aurora/tree/main/helm) · [Source](https://github.com/aurorallm/aurora)
+
+---
+
+## Kubernetes (Helm)
+
+Deploy on any Kubernetes cluster with the [Helm chart](https://github.com/aurorallm/aurora/tree/main/helm).
+
+```bash
+# Quick dev — Groq, no Redis, no auth
+helm install aurora ./helm \
+  --namespace aurora --create-namespace \
+  --set image.tag=1.0.25 \
+  --set providers.groq.apiKey="gsk_your_key_here" \
+  --set providers.groq.enabled=true \
+  --set redis.enabled=false \
+  --set auth.masterKey=""
+```
+
+```bash
+# Production — multiple providers, auth, Redis
+helm upgrade --install aurora ./helm \
+  --namespace aurora --create-namespace \
+  --set image.tag=1.0.25 \
+  --set auth.masterKey="your-secure-key" \
+  --set providers.openai.apiKey="sk-..." \
+  --set providers.openai.enabled=true \
+  --set providers.anthropic.apiKey="sk-ant-..." \
+  --set providers.anthropic.enabled=true \
+  --set redis.enabled=true
+```
+
+**Full Helm documentation:** [helm/README.md](https://github.com/aurorallm/aurora/tree/main/helm/README.md)
 
 ---
 
@@ -91,51 +108,23 @@ The dashboard is at `http://localhost:8080/admin/dashboard`.
 Aurora supports enterprise-grade deployments for teams running production AI systems at scale.
 In addition to private networking, custom security controls, and governance, **Aurora Enterprise** unlocks advanced capabilities including SSO, RBAC, tenant isolation, budget enforcement, compliance workflows, and production support.
 
-The Enterprise edition is a separate distribution with a signed license — it cannot be enabled by editing files in this tree.
+The Enterprise edition is a separate distribution with a signed license.
 
-<a href="./comparison.png">
-  <img src="./comparison.png" alt="OSS vs Enterprise comparison" width="100%" style="border-radius:12px;margin:16px 0;">
+<a href="https://raw.githubusercontent.com/aurorallm/aurora/main/docs/assets/comparison.png">
+  <img src="https://raw.githubusercontent.com/aurorallm/aurora/main/docs/assets/comparison.png" alt="OSS vs Enterprise comparison" width="100%" style="border-radius:12px;margin:16px 0;">
 </a>
 
 <div align="center">
-  <a href="https://discord.com/invite/YJPrfR9uh">
-    <img src="https://img.shields.io/badge/Contact%20Us-Enterprise-5865F2?style=for-the-badge&logo=discord&logoColor=white" alt="Contact Us" width="200" style="margin-top:5px;"/>
+  <a href="mailto:team.auroragate@gmail.com?subject=Aurora%20Enterprise%20Inquiry" style="display:block;margin-top:5px;">
+    <img src="https://img.shields.io/badge/Email%20Us-Enterprise-5865F2?style=for-the-badge&logo=gmail&logoColor=white" alt="Email Aurora Enterprise" width="200"/>
   </a>
 </div>
 
 ---
 
-## Key Features
-
-### Core Infrastructure
-
-- **Unified Interface** — Single OpenAI- and Anthropic-compatible API for all providers
-- **Multi-Provider Support** — 15+ providers auto-discovered from env vars (OpenAI, Anthropic, Gemini, Groq, DeepSeek, and more)
-- **Automatic Fallbacks** — Seamless failover with fallback routing and combo models
-- **Model Discovery** — Providers and models auto-detected from environment variables
-
-### Advanced Features
-
-- **Admin Dashboard** — Web UI for managing providers, API keys, models, and combos
-- **Semantic Caching** — Intelligent response caching based on semantic similarity to reduce costs and latency
-- **Anthropic Ingress** — Native Anthropic-format `/v1/messages` endpoint
-- **Audit Logging** — Records every request/response with retention
-- **Guardrails** — Content safety filters (system prompt, regex, PII redaction)
-
-### Developer Experience
-
-- **Zero-Config Startup** — Start immediately with auto-detected providers
-- **Prometheus Metrics** — Go runtime + request metrics at `/metrics`
-- **Swagger UI** — Interactive API docs at `/swagger/index.html`
-- **SQLite Storage** — Zero-config embedded database, no setup needed
-- **CLI Tools** — Admin configuration via CLI commands
-
----
 
 ## Repository Structure
 
-Aurora uses a modular architecture for maximum flexibility:
-
 ```text
 aurora/
 ├── apps/                  # Application entrypoints

package/package.json

@@ -1,5 +1,5 @@
 {
-    "version":  "1.0.24",
+    "version":  "1.0.26",
     "author":  "Aurora Gateway",
     "license":  "Apache-2.0",
     "files":  [