Fully autonomous AI Agents system capable of performing complex penetration testing tasks
Go ai-agentsai-security-toolanthropicautonomous-agentsgolanggptgraphqlmulti-agent-systemoffensive-securityopen-sourceopenaipenetration-testingpenetration-testing-toolsreactsecurity-automationsecurity-testingsecurity-toolsself-hosted
package main
import (
"database/sql"
"fmt"
"strings"
"pentagi/pkg/terminal"
)
const (
testText = "This is a test text for embedding"
testTexts = "This is a test text for embedding\nThis is another test text for embedding"
)
// test checks connectivity to the database and tests the embedder.
func (t *Tester) test() error {
terminal.Info("Testing connection to PostgreSQL database... ")
err := t.conn.Ping(t.ctx)
if err != nil {
terminal.Error("FAILED")
return fmt.Errorf("database connection test failed: %w", err)
}
terminal.Success("OK")
terminal.Info("Testing pgvector extension... ")
var result string
err = t.conn.QueryRow(t.ctx, "SELECT extname FROM pg_extension WHERE extname = 'vector'").Scan(&result)
if err != nil {
if err == sql.ErrNoRows {
terminal.Error("FAILED")
return fmt.Errorf("pgvector extension is not installed")
}
terminal.Error("FAILED")
return fmt.Errorf("failed to check pgvector extension: %w", err)
}
terminal.Success("OK")
terminal.Info("Testing embedding table existence... ")
var tableExists bool
err = t.conn.QueryRow(t.ctx,
"SELECT EXISTS (SELECT FROM information_schema.tables WHERE table_name = $1)",
t.embeddingTableName).Scan(&tableExists)
if err != nil {
terminal.Error("FAILED")
return fmt.Errorf("failed to check embedding table: %w", err)
}
if !tableExists {
terminal.Error("FAILED")
return fmt.Errorf("embedding table '%s' does not exist", t.embeddingTableName)
}
terminal.Success("OK")
terminal.Info("Testing embedder with single query... ")
if !t.embedder.IsAvailable() {
terminal.Error("FAILED")
return fmt.Errorf("embedder is not available")
}
embedVector, err := t.embedder.EmbedQuery(t.ctx, testText)
if err != nil {
terminal.Error("FAILED")
return fmt.Errorf("embedder test failed: %w", err)
}
if len(embedVector) == 0 {
terminal.Error("FAILED")
return fmt.Errorf("embedder returned empty vector")
}
terminal.Success(fmt.Sprintf("OK (%d dimensions)", len(embedVector)))
terminal.Info("Testing embedder with multiple documents... ")
texts := strings.Split(testTexts, "\n")
embedVectors, err := t.embedder.EmbedDocuments(t.ctx, texts)
if err != nil {
terminal.Error("FAILED")
return fmt.Errorf("embedder multi-text test failed: %w", err)
}
if len(embedVectors) != len(texts) {
terminal.Error("FAILED")
return fmt.Errorf("embedder returned wrong number of vectors: got %d, expected %d",
len(embedVectors), len(texts))
}
if len(embedVectors[0]) == 0 || len(embedVectors[1]) == 0 {
terminal.Error("FAILED")
return fmt.Errorf("embedder returned empty vectors")
}
terminal.Success(fmt.Sprintf("OK (%d documents, %d dimensions each)",
len(embedVectors), len(embedVectors[0])))
if t.verbose {
terminal.PrintHeader("\nVerbose output:")
terminal.PrintKeyValue("Embedding provider", t.cfg.EmbeddingProvider)
terminal.PrintKeyValueFormat("Vector dimensions", "%d", len(embedVector))
// Display a sample of vector values for inspection
vectorSample := embedVector
if len(vectorSample) > 5 {
vectorSample = vectorSample[:5]
}
terminal.PrintKeyValue("First values of test vector",
fmt.Sprintf("%v", vectorSample))
}
terminal.Success("\nAll tests passed successfully!")
return nil
}