Fix false positives in agent guardrails

Skip empty fingerprints in SystemPromptLeakGuardrail to prevent blank
values from flagging every message. Replace overly broad "sk-" pattern
in SensitiveDataGuardrail with specific LLM provider prefixes
("sk-proj-" for OpenAI, "sk-ant-" for Anthropic) to avoid false
positives on common words like "risk-based" or "task-management".

Signed-off-by: Bryan Frimin <bryan@getprobo.com>

pkg/agent/guardrail/sensitive_data_test.go

@@ -66,6 +66,8 @@ func TestSensitiveDataGuardrail_Check(t *testing.T) {
 
 		// LLM provider keys
 		{"openai key", "The API key is sk-proj-abc123", true},
+		{"anthropic key", "Key: sk-ant-api03-abc123", true},
+		{"sk prefix not a false positive", "This is a risk-based approach to task-management.", false},
 
 		// JWT tokens
 		{"jwt token", "Token: eyJhbGciOiJIUzI1NiJ9.payload.sig", true},