Tencent/WeKnora · error

chat model not available for web_fetch summary

Error message

chat model not available for web_fetch summary

What it means

Dependency guard in processWithLLM: the chat model handle is nil (LLM not configured for this web_fetch instance), so fetched page content cannot be summarized and the per-item processing step fails rather than nil-panicking.

Source

Thrown at internal/agent/tools/web_fetch.go:237

		Data: map[string]interface{}{
			"results":          aggregated,
			"count":            len(aggregated),
			"successful_count": successCount,
			"failed_count":     failedCount,
			"skipped_count":    skippedCount,
			"all_failed":       allFailed,
			"display_type":     "web_fetch_results",
		},
	}
	if allFailed {
		toolResult.Error = "all page fetches failed"
	}
	return toolResult
}

func (t *WebFetchTool) processWithLLM(ctx context.Context, item WebFetchItem, content string) (string, error) {
	if t.chatModel == nil {
		return "", fmt.Errorf("chat model not available for web_fetch summary")
	}
	messages := []chat.Message{
		{
			Role:    "system",
			Content: "Answer the request from the supplied web page text. Never fabricate information that is absent from the page.",
		},
		{
			Role:    "user",
			Content: fmt.Sprintf("User request:\n%s\n\nWeb page content:\n%s", item.Prompt, content),
		},
	}
	modelCtx := types.WithLLMCallMetadata(ctx, "web_fetch_summary", "")
	response, err := t.chatModel.Chat(modelCtx, messages, &chat.ChatOptions{Temperature: 0.3, MaxTokens: 1024})
	if err != nil {
		return "", err
	}
	return strings.TrimSpace(response.Content), nil
}

View on GitHub (pinned to 988cbb0330)

Solutions

  1. Configure a chat model for the web_fetch tool
  2. Skip LLM summarization and return raw content when no model is available
  3. Report summarization unavailability distinctly from fetch failure
Defensive patterns

Strategy: fallback

When it happens

Trigger: Thrown at internal/agent/tools/web_fetch.go:237 when the library encounters an invalid state.

Common situations: See trigger scenarios.


AI-assisted analysis of Tencent/WeKnora@988cbb0330 (2026-09-02). Data as JSON: /api/errors/b06299f76a9e12b1. Report an issue: GitHub.