Initial commit: LlamaLink Go rewrite

Complete rewrite from Python/FastAPI to Go/Gin:
- Go backend: auth (API keys + bcrypt), llama.cpp subprocess manager,
  hot-swap multi-model, rate limiting, quota system, webhooks
- Vue 3 SPA admin panel (src/) with Tailwind CSS
- Deployment: Docker multi-stage, docker-compose, nginx, systemd
- GORM/SQLite models: ApiKey, Model, UsageLog, Quota, Webhook
- REST API: /api/v1/admin/* (keys, models, chat, usage, health)
- Embedded frontend via go:embed (build output at web/dist/)

Removed legacy Python artifacts (app/, tests/, pyproject.toml, etc.)
This commit is contained in:
2026-07-30 10:58:55 -04:00
commit 4c9ed3c24b
52 changed files with 5105 additions and 0 deletions
+156
View File
@@ -0,0 +1,156 @@
package middleware
import (
"net/http"
"strings"
"github.com/gin-gonic/gin"
"github.com/google/uuid"
"github.com/llamalink/llamalink/internal/auth"
"github.com/llamalink/llamalink/internal/db"
)
const (
ApiKeyCtx = "api_key"
ApiKeyIDCtx = "api_key_id"
)
func APIKeyAuth(authService *auth.Service) gin.HandlerFunc {
return func(c *gin.Context) {
authHeader := c.GetHeader("Authorization")
if authHeader == "" {
c.AbortWithStatusJSON(http.StatusUnauthorized, gin.H{
"error": gin.H{
"code": "invalid_api_key",
"message": "Authorization header required",
},
})
return
}
token := strings.TrimPrefix(authHeader, "Bearer ")
if token == authHeader {
c.AbortWithStatusJSON(http.StatusUnauthorized, gin.H{
"error": gin.H{
"code": "invalid_api_key",
"message": "Bearer token required",
},
})
return
}
apiKey, err := authService.Validate(token)
if err != nil {
code := "invalid_api_key"
status := http.StatusUnauthorized
if err == auth.ErrKeyRevoked || err == auth.ErrKeyExpired {
code = "api_key_revoked"
status = http.StatusUnauthorized
}
c.AbortWithStatusJSON(status, gin.H{
"error": gin.H{
"code": code,
"message": err.Error(),
},
})
return
}
c.Set(ApiKeyCtx, apiKey)
c.Set(ApiKeyIDCtx, apiKey.ID)
c.Next()
}
}
func RequireScope(authService *auth.Service, scope string) gin.HandlerFunc {
return func(c *gin.Context) {
apiKey, exists := c.Get(ApiKeyCtx)
if !exists {
c.AbortWithStatusJSON(http.StatusUnauthorized, gin.H{
"error": gin.H{
"code": "invalid_api_key",
"message": "Authentication required",
},
})
return
}
key := apiKey.(*db.ApiKey)
if !authService.HasScope(key, scope) {
c.AbortWithStatusJSON(http.StatusForbidden, gin.H{
"error": gin.H{
"code": "insufficient_scope",
"message": "API key lacks required scope: " + scope,
},
})
return
}
c.Next()
}
}
func AdminOnly() gin.HandlerFunc {
return func(c *gin.Context) {
apiKey, exists := c.Get(ApiKeyCtx)
if !exists {
c.AbortWithStatusJSON(http.StatusUnauthorized, gin.H{
"error": gin.H{
"code": "unauthorized",
"message": "Admin access required",
},
})
return
}
key := apiKey.(*db.ApiKey)
if !key.IsAdmin {
c.AbortWithStatusJSON(http.StatusForbidden, gin.H{
"error": gin.H{
"code": "unauthorized",
"message": "Admin access required",
},
})
return
}
c.Next()
}
}
func AdminTokenAuth(token string) gin.HandlerFunc {
return func(c *gin.Context) {
adminToken := c.GetHeader("X-Admin-Token")
if adminToken == "" {
adminToken = c.GetHeader("Authorization")
adminToken = strings.TrimPrefix(adminToken, "Bearer ")
}
if adminToken != token {
c.AbortWithStatusJSON(http.StatusUnauthorized, gin.H{
"error": gin.H{
"code": "unauthorized",
"message": "Invalid admin token",
},
})
return
}
c.Next()
}
}
func GetAPIKeyID(c *gin.Context) uuid.UUID {
id, _ := c.Get(ApiKeyIDCtx)
return id.(uuid.UUID)
}
func GetAPIKey(c *gin.Context) *db.ApiKey {
key, _ := c.Get(ApiKeyCtx)
if key == nil {
return nil
}
return key.(*db.ApiKey)
}
+85
View File
@@ -0,0 +1,85 @@
package middleware
import (
"net/http"
"sync"
"time"
"github.com/gin-gonic/gin"
"golang.org/x/time/rate"
)
type RateLimiter struct {
visitors map[string]*visitor
mu sync.RWMutex
rate rate.Limit
burst int
}
type visitor struct {
limiter *rate.Limiter
lastSeen time.Time
}
func NewRateLimiter(requestsPerMinute int) *RateLimiter {
rl := &RateLimiter{
visitors: make(map[string]*visitor),
rate: rate.Limit(float64(requestsPerMinute) / 60.0),
burst: requestsPerMinute / 10,
}
if requestsPerMinute < 10 {
rl.burst = 1
}
// Cleanup old visitors
go rl.cleanup()
return rl
}
func (rl *RateLimiter) cleanup() {
ticker := time.NewTicker(5 * time.Minute)
for range ticker.C {
rl.mu.Lock()
for ip, v := range rl.visitors {
if time.Since(v.lastSeen) > 10*time.Minute {
delete(rl.visitors, ip)
}
}
rl.mu.Unlock()
}
}
func (rl *RateLimiter) getVisitor(ip string) *rate.Limiter {
rl.mu.Lock()
defer rl.mu.Unlock()
v, exists := rl.visitors[ip]
if !exists {
v = &visitor{
limiter: rate.NewLimiter(rl.rate, rl.burst),
lastSeen: time.Now(),
}
rl.visitors[ip] = v
}
v.lastSeen = time.Now()
return v.limiter
}
func RateLimitMiddleware(rl *RateLimiter) gin.HandlerFunc {
return func(c *gin.Context) {
ip := c.ClientIP()
if !rl.getVisitor(ip).Allow() {
c.AbortWithStatusJSON(http.StatusTooManyRequests, gin.H{
"error": gin.H{
"code": "rate_limit_exceeded",
"message": "Too many requests",
"retry_after_seconds": 60,
},
})
return
}
c.Next()
}
}