vxcontrol/pentagi · error

failed to test agent: %w

Error message

failed to test agent: %w

What it means

TestAgent runs tester.TestProvider against the temporary provider, limited to the requested agent type via tester.WithAgentTypes. If the test harness itself returns an error (setup failure, context cancellation, harness-internal error — not per-request model failures, which are recorded in results), it is wrapped as "failed to test agent: %w".

Source

Thrown at backend/pkg/providers/providers.go:900

	}

	// Create temporary provider for testing using existing provider logic
	providerName := provider.ProviderName("test-provider")
	tempProvider, err := pc.buildProviderFromConfig(prvtype, providerName, patchedConfig)
	if err != nil {
		return result, fmt.Errorf("failed to create provider for testing: %w", err)
	}

	// Run tests for specific agent type only
	results, err := tester.TestProvider(
		ctx,
		tempProvider,
		tester.WithAgentTypes(agentType),
		tester.WithVerbose(false),
		tester.WithParallelWorkers(defaultTestParallelWorkersNumber),
	)
	if err != nil {
		return result, fmt.Errorf("failed to test agent: %w", err)
	}

	// Extract results for the specific agent type
	switch agentType {
	case pconfig.OptionsTypeSimple:
		result = results.Simple
	case pconfig.OptionsTypeSimpleJSON:
		result = results.SimpleJSON
	case pconfig.OptionsTypePrimaryAgent:
		result = results.PrimaryAgent
	case pconfig.OptionsTypeAssistant:
		result = results.Assistant
	case pconfig.OptionsTypeGenerator:
		result = results.Generator
	case pconfig.OptionsTypeRefiner:
		result = results.Refiner
	case pconfig.OptionsTypeAdviser:
		result = results.Adviser

View on GitHub (pinned to ea665308ba)

Solutions

1. Inspect the wrapped inner error for timeout vs harness-internal cause
2. Increase the context timeout or check network reachability to the LLM endpoint
3. Confirm the agent type passed to WithAgentTypes is supported by the tester package
4. Retry once to rule out a transient provider outage before deeper debugging

Example fix

// before
ctx := context.Background() // no deadline; hangs then fails
// after
ctx, cancel := context.WithTimeout(context.Background(), 2*time.Minute)
defer cancel()
results, err := tester.TestProvider(ctx, tempProvider, tester.WithAgentTypes(agentType))

ctx, cancel := context.WithTimeout(ctx, 2*time.Minute)
defer cancel() // ensure adequate deadline before the test call

result, err := ctrl.TestAgent(ctx, prvtype, agentType, cfg)
if err != nil {
    if errors.Is(err, context.DeadlineExceeded) || strings.Contains(err.Error(), "failed to test agent") {
        return fmt.Errorf("agent test timed out or harness failed; retry or increase timeout: %w", err)
    }
    return err
}

When it happens

Trigger: tester.TestProvider returns an error: invalid test options, an unsupported agent type inside the tester, context deadline/cancellation during the test run, or a harness-level setup error (e.g. cannot enumerate tools for the agent type).

Common situations: Test context times out while the model call runs; passing an agent type the tester does not support; parallel worker configuration invalid; underlying LLM endpoint hangs until ctx deadline.

Related errors

• failed to test provider: %w
• edit_file was called but did not produce the requested chang
• failed to switch provider: %w
• flow %d input processing timeout: %w
• task stop timeout

AI-assisted analysis of vxcontrol/pentagi@ea665308ba (2026-09-01). Data as JSON: /api/errors/6c188125522a0107. Report an issue: GitHub.