internal/core/config/config_test.go (1)

160-256: Well-structured table-driven tests — consider adding a whitespace-only value case.

All 5 required fields and the partial-config scenario are covered. Validate() explicitly calls strings.TrimSpace, but no test case exercises a field value that is non-empty yet all-whitespace (e.g., URL: " "). Adding one case would confirm that strings.TrimSpace behavior is intentional and tested.

♻️ Proposed additional test case

 		{
 			name: "partial config",
 			cfg: Config{
 				Qdrant: QdrantConfig{
 					URL: "https://example.qdrant.io:6334",
 				},
 			},
 			wantErr: "config validation failed: qdrant.api_key is empty (check QDRANT_API_KEY environment variable)",
 		},
+		{
+			name: "whitespace-only qdrant url",
+			cfg: func() Config {
+				cfg := baseConfig
+				cfg.Qdrant.URL = "   "
+				return cfg
+			}(),
+			wantErr: "config validation failed: qdrant.url is empty (check QDRANT_URL environment variable)",
+		},

🤖 Prompt for AI Agents

Verify each finding against the current code and only fix it if needed.

In `@internal/core/config/config_test.go` around lines 160 - 256, Add a
whitespace-only value test to TestConfigValidate to exercise strings.TrimSpace
behavior: create a new test case in the tests slice that copies baseConfig, sets
a field to a whitespace-only string (e.g., cfg.Qdrant.URL = "   " or
cfg.Embedding.APIKey = "   "), call cfg.Validate() and assert it returns the
same error string as the empty-value case (e.g., "config validation failed:
qdrant.url is empty (check QDRANT_URL environment variable)"); this verifies
Validate() treats all-whitespace as empty.

internal/core/config/config.go (1)

190-215: Consider collecting all validation errors before returning.

The current fail-fast approach (returning on the first empty field) requires a fix-and-restart cycle for each missing field. When multiple required fields are absent on a first deployment, users face sequential error discovery. Aggregating all failures into a single error would improve the startup experience.

♻️ Proposed refactor to collect all missing fields

 func (c *Config) Validate() error {
 	requiredFields := []struct {
 		name   string
 		envVar string
 		value  string
 	}{
 		{name: "qdrant.url", envVar: "QDRANT_URL", value: c.Qdrant.URL},
 		{name: "qdrant.api_key", envVar: "QDRANT_API_KEY", value: c.Qdrant.APIKey},
 		{name: "qdrant.collection", envVar: "QDRANT_COLLECTION", value: c.Qdrant.Collection},
 		{name: "embedding.api_key", envVar: "EMBEDDING_API_KEY", value: c.Embedding.APIKey},
 		{name: "llm.api_key", envVar: "LLM_API_KEY", value: c.LLM.APIKey},
 	}

+	var missing []string
 	for _, field := range requiredFields {
 		if strings.TrimSpace(field.value) == "" {
-			return fmt.Errorf(
-				"config validation failed: %s is empty (check %s environment variable)",
-				field.name,
-				field.envVar,
-			)
+			missing = append(missing, fmt.Sprintf("%s (check %s environment variable)", field.name, field.envVar))
 		}
 	}

+	if len(missing) > 0 {
+		return fmt.Errorf("config validation failed: missing required fields: %s", strings.Join(missing, "; "))
+	}
+
 	return nil
 }

🤖 Prompt for AI Agents

Verify each finding against the current code and only fix it if needed.

In `@internal/core/config/config.go` around lines 190 - 215, Change
Config.Validate to collect all missing required fields instead of returning on
the first error: iterate the requiredFields slice (the local variable in
Validate that contains entries like {name: "qdrant.url", envVar: "QDRANT_URL",
value: c.Qdrant.URL}), append a descriptive message for each empty value (use
strings.TrimSpace(field.value) == "") to a list, and after the loop return a
single aggregated error that joins all messages (or nil if none). Keep the same
message format per field ("<name> is empty (check <envVar> environment
variable)") and ensure the function still returns nil when there are no missing
fields.