quick

Execute quick ad-hoc tasks via a towline-executor subagent with PLAN.md and SUMMARY.md artifacts.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/SienkLogic/towline --skill quick-sienklogic
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: quick
Source: https://github.com/SienkLogic/towline/tree/main/plugins/dev/skills/quick
Command: npx skills add https://github.com/SienkLogic/towline --skill quick-sienklogic

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Quick tasks often run without formal planning or tracking, leading to undocumented results and inconsistent outcomes. This Skill provides a lightweight, auditable workflow that executes small, well-scoped tasks via a dedicated subagent, while automatically creating planning artifacts and a concise result report.

Core Features & Use Cases

  • Spawns a single Task(subagent_type: "dev:towline-executor") to perform the work.
  • Creates and uses a dedicated .planning/quick/{NNN}-{slug}/ directory with PLAN.md and SUMMARY.md, and updates STATE.md for progress tracking.
  • Enforces an auditable, atomic commit model and a final results report to ensure traceability for quick tasks.

Quick Start

Describe the task you need done and Towline will create a quick task plan, spawn a towline-executor subagent, and report results.

Frequently Asked Questions about quick

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

FAQPage Schema
How do I automate quick ad-hoc tasks with tracked planning artifacts?

Automating quick ad-hoc tasks with tracked planning artifacts involves spawning a dedicated towline-executor subagent to execute well-scoped work while automatically generating PLAN.md and SUMMARY.md files to ensure auditable, documented outcomes.

What is the best way to track small, well-defined tasks in a Towline project?

Tracking small tasks in a Towline project is best handled by creating a dedicated .planning/quick/ directory, updating STATE.md for progress monitoring, and enforcing an atomic commit model for traceable results.

How do I execute a single subagent run for a quick task?

To execute a single subagent run for a quick task, you describe the task to initiate the creation of a plan, which then spawns a single Task with subagent_type "dev:towline-executor" to perform the implementation without direct code writing.

Can I use this approach for tasks that require multiple subagent runs?

This approach is not suitable for tasks requiring multiple subagent runs, as it is specifically designed for small, well-scoped tasks that can be completed atomically within a single subagent execution cycle.

Why do quick tasks need formal planning and result reports?

Quick tasks need formal planning and result reports because running them without tracking leads to undocumented results and inconsistent outcomes, making automated PLAN.md and SUMMARY.md artifacts essential for auditability.