loom-plan-writer

Generate and validate Loom execution plans with mandatory stages and verification fields.

53|1|Updated Dec 18, 2025
One-click install
npx skills add https://github.com/cosmix/loom --skill loom-plan-writer
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: loom-plan-writer
Source: https://github.com/cosmix/loom/tree/main/skills/loom-plan-writer
Command: npx skills add https://github.com/cosmix/loom --skill loom-plan-writer

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill is essential for creating structured execution plans for the Loom agent orchestration system, ensuring that plans are correctly formatted, include necessary verification steps, and leverage parallelization effectively.

Core Features & Use Cases

  • Mandatory Plan Structure: Enforces the inclusion of knowledge-bootstrap and integration-verify stages.
  • Parallelization Strategy: Guides the optimal use of parallel subagents and concurrent stages for maximum throughput.
  • Verification Enforcement: Ensures that all stages include necessary verification fields (truths, artifacts, wiring).
  • Use Case: When you need to define a complex software development task for Loom, this skill will help you structure the entire process, from initial knowledge gathering to final verification, ensuring all steps are clearly defined and validated.

Quick Start

Use the loom-plan-writer skill to create a new Loom execution plan for a feature.

Frequently Asked Questions about loom-plan-writer

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

FAQPage Schema
How do I create an execution plan for agent orchestration?

To create an execution plan for agent orchestration, generate a structured plan with mandatory knowledge-bootstrap and integration-verify stages. The plan must define stage dependencies, sandbox configurations, and working directories to facilitate structured subagent execution.

How do I enforce verification steps in a multi-agent workflow?

To enforce verification steps in a multi-agent workflow, include mandatory verification fields like truths, artifacts, and wiring in your plan. The integration-verify stage ensures all stages are validated and correctly wired before execution completes.

What is the best way to optimize parallel subagent execution?

The best way to optimize parallel subagent execution is to design plans that leverage concurrent stages. Structuring stage dependencies effectively allows the orchestration system to maximize throughput by running independent subagents simultaneously.

Do I need a knowledge-bootstrap stage for every agent execution plan?

Yes, a knowledge-bootstrap stage is mandatory for every agent execution plan. It ensures the orchestrator gathers necessary initial knowledge before proceeding to parallel subagent stages and final integration verification.

Can I define sandbox configurations and working directories in an orchestration plan?

Yes, you can define sandbox configurations and working directories within the orchestration plan. Specifying these environments ensures that concurrent stages and parallel subagents execute in isolated, correctly configured workspaces.

How do I validate stage dependencies in a concurrent execution workflow?

You validate stage dependencies in a concurrent execution workflow by enforcing strict plan structures. The integration-verify stage checks that all defined dependencies, artifacts, and wiring are correctly configured before the workflow completes.