tasks

Create and manage subtasks for deferred execution of long-running tasks.

10|Updated Feb 17, 2026
One-click install
npx skills add https://github.com/istota-project/istota --skill tasks-istota-project
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: tasks
Source: https://github.com/istota-project/istota/tree/main/src/istota/skills/tasks
Command: npx skills add https://github.com/istota-project/istota --skill tasks-istota-project

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Orchestrates the creation and scheduling of subtasks that run after a parent task completes, enabling better task decomposition and reliability.

Core Features & Use Cases

  • Subtask creation: define subtasks with prompts and optional fields like conversation_token and priority.
  • Deferred execution: subtasks are processed after the parent task, ensuring ordered work.
  • Use case: break a long task into smaller steps to improve visibility and resilience.

Quick Start

Create a JSON file at the designated deferred directory to define subtasks for your task.

Frequently Asked Questions about tasks

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I queue subtasks for deferred execution after a parent task completes?

Queue subtasks for deferred execution by creating a JSON file in the designated deferred directory, defining fields like prompt, conversation_token, and priority to ensure ordered processing after the parent task.

What environment variables are required to schedule async subtasks?

Scheduling async subtasks requires the ISTOTA_DEFERRED_DIR for output, along with ISTOTA_TASK_ID and ISTOTA_USER_ID context variables to properly route and manage the deferred task queue.

How does prioritization work when breaking long-running tasks into smaller prompts?

Prioritization works by assigning a priority field within the JSON structure of each subtask, allowing the deferred execution queue to process smaller prompts in a specified order once the parent task finishes.

Can I pass conversation context to deferred subtasks in an ai-workflow?

Yes, you can pass conversation context to deferred subtasks by using the optional conversation_token field in the JSON structure, maintaining context continuity across the async task queue.

Why use subtask decomposition for complex async workflows instead of single prompts?

Subtask decomposition improves visibility and resilience by splitting complex tasks into smaller, ordered steps, ensuring post-parent-task processing can handle individual failures without losing the entire workflow's progress.

What's the best way to structure JSON for a task queue of deferred subtasks?

Structure JSON for a task queue by defining an array of subtask objects, each containing a required prompt field and optional conversation_token and priority fields, saved directly to the deferred directory.