vxcontrol/pentagi · warning
failed to summarize task list: %w
Error message
failed to summarize task list: %w
What it means
Thrown in buildTasksList (backend/pkg/tools/flow_manager.go:200) when the assembled task-list string exceeds taskListLimit (32 KB) and the configured summarizer fails while compressing it (input pre-truncated to summarizationLimit, 128 KB). The task data was fully rendered; only the final LLM compression of the list failed.
Source
Thrown at backend/pkg/tools/flow_manager.go:200
return "", fmt.Errorf("failed to get task input: %w", err)
}
fmt.Fprintf(sb, " Input: %s\n", input)
}
if task.Result != "" {
result, err := t.getResultText(ctx, task.Result)
if err != nil {
return "", fmt.Errorf("failed to get task result: %w", err)
}
fmt.Fprintf(sb, " Result: %s\n", result)
}
}
}
taskList := sb.String()
if t.summarizer != nil && len(taskList) > taskListLimit {
taskList, err = t.summarizer(ctx, truncateText(taskList, summarizationLimit))
if err != nil {
return "", fmt.Errorf("failed to summarize task list: %w", err)
}
}
return taskList, nil
}
func (t *flowStatusTool) buildSubtasksList(ctx context.Context, taskID *int64, verbose bool) (string, error) {
var subtasks []database.Subtask
var err error
if taskID != nil && *taskID > 0 {
subtasks, err = t.db.GetFlowTaskSubtasks(ctx, database.GetFlowTaskSubtasksParams{
FlowID: t.flowID,
TaskID: *taskID,
})
} else {
subtasks, err = t.db.GetFlowSubtasks(ctx, t.flowID)
}View on GitHub (pinned to ea665308ba)
Solutions
- Check and fix the summarizer provider (credentials, quota, endpoint) per the wrapped error, then retry.
- Reduce detail: call with verbose=false or paginate by filtering tasks, keeping output below 32 KB.
- Extend the summarizer timeout or use a model with a larger context window for the 128 KB input.
- Disable the summarizer so the raw (already size-bounded by callers) list is returned truncated.
Example fix
// before
taskList, err = t.summarizer(ctx, truncateText(taskList, summarizationLimit))
if err != nil {
return "", fmt.Errorf("failed to summarize task list: %w", err)
}
// after: fall back to the non-summarized list
rendered, serr := t.summarizer(ctx, truncateText(taskList, summarizationLimit))
if serr != nil {
log.Warn("task list summarization failed", "err", serr)
return truncateText(taskList, taskListLimit), nil
}
taskList = rendered Defensive patterns
Strategy: fallback
Validate before calling
// estimate output size and avoid oversized verbose dumps
if estimatedTaskListSize(flowID, verbose) > taskListLimit && !summarizerHealthy(ctx) {
args.Verbose = false
} Try / catch
// Go: accept the unsanitized list if compression fails
out, err := tool.Handle(ctx, "get_flow_status", verboseTasksArgs)
if err != nil && strings.Contains(err.Error(), "failed to summarize task list") {
out, err = tool.Handle(ctx, "get_flow_status", nonVerboseTasksArgs) // smaller, no LLM
} Prevention
- Avoid verbose=true on very large flows; fetch specific tasks instead.
- Choose a summarizer model with sufficient context and latency budget.
- Alert on summarizer error rate; disable summarization gracefully when unhealthy.
- Archive or prune old tasks to keep task lists below the 32 KB limit.
When it happens
Trigger: get_flow_status with detail='tasks' on a flow with many tasks (especially verbose=true with inputs/results), producing >32 KB output, combined with a summarizer failure — provider error, timeout, 429, cancelled context.
Common situations: Very long-running flows with dozens of tasks; verbose dumps pushing output past 32 KB; summarizer provider quota exhausted after the whole run; local summarizer model offline; caller context deadline exceeded during compression.
Related errors
- failed to summarize summary: %w
- failed to get active task input: %w
- failed to get active subtask description: %w
- failed to get task input: %w
- failed to get task result: %w
AI-assisted analysis of vxcontrol/pentagi@ea665308ba (2026-09-01).
Data as JSON: /api/errors/aa42f00066bac824.
Report an issue: GitHub.