myco:author-harness-task

Automate Phase-based TaskDefinition authoring for the Myco agent harness.

12|2|Updated Mar 15, 2026
One-click install
npx skills add https://github.com/goondocks-co/myco --skill myco-author-harness-task
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: myco:author-harness-task
Source: https://github.com/goondocks-co/myco/tree/main/.agents/skills/author-harness-task
Command: npx skills add https://github.com/goondocks-co/myco --skill myco-author-harness-task

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Designers need a structured method to author, configure, and debug phased executor tasks for the Myco agent harness, enabling reliable task lifecycles from discovery to execution.

Core Features & Use Cases

  • TaskDefinition scaffolding: publish a typed TaskDefinition with phases and triggers.
  • Phase workflow design: design 'discover' and 'write' phases with budgets and tools, plus model selection per phase.
  • Scheduling, gating, and observability: configure triggers, session gating, and inspect agent_runs for diagnostics.

Quick Start

Create a new TypeScript task file exporting a TaskDefinition in the tasks directory and restart the daemon to load it.

Frequently Asked Questions about myco:author-harness-task

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

FAQPage Schema
How do I define a phased task for the Myco agent harness?

You define a phased task by creating a TypeScript file exporting a typed TaskDefinition with configured phases and triggers in the designated tasks directory. This registers the structured task lifecycle for the Myco agent harness.

How does phase workflow design work for budget calibration and model selection?

Phase workflow design lets you configure specific 'discover' and 'write' phases with individual budgets, tool surfaces, and model selection. This ensures each phase of the TaskDefinition executes within its calibrated resource limits.

Do I need a running daemon to test a new TaskDefinition?

Yes, a running daemon is required to load and test a new TaskDefinition. After placing the TypeScript task file in the correct directory, restarting the daemon registers the configuration so you can validate the phased execution.

What is the best way to configure session gating and triggers for agent tasks?

The best way to configure session gating and triggers is within the TaskDefinition scaffold. This structured method maps phase workflows to specific triggers and gates reliable task lifecycles from discovery to execution.

Why are my phased executor tasks failing during diagnostics and how do I inspect them?

Phased executor tasks failing during diagnostics can be inspected by checking agent_runs. The TaskDefinition framework provides observability into these runs to help debug phase workflows, budgets, and tool surfaces.