speckit-conduct-run

Execute spec-kit phases by parsing instructions and delegating steps to sub-agents.

2|Updated Jan 6, 2026
One-click install
npx skills add https://github.com/NUMU-IO/NUMU-api --skill speckit-conduct-run
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: speckit-conduct-run
Source: https://github.com/NUMU-IO/NUMU-api/tree/main/.agents/skills/speckit-conduct-run
Command: npx skills add https://github.com/NUMU-IO/NUMU-api --skill speckit-conduct-run

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill removes the manual coordination burden of running a spec-kit phase by reading the phase instructions, orchestrating sub-agents one step at a time, and producing a concise summary of the results.

Core Features & Use Cases

  • Resolves the requested phase and loads the matching local instruction file.
  • Breaks the phase into discrete steps and delegates each step sequentially to specialized sub-agents.
  • Tracks files created or modified during execution and reads them back for a content-rich final summary.
  • Use it when you need to execute a spec-kit workflow such as specification, planning, or task execution without mixing steps or losing context.

Quick Start

Use the speckit conduct run skill to execute the requested phase for the current spec-kit project and summarize the outcome.

Frequently Asked Questions about speckit-conduct-run

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

FAQPage Schema
How do I automate spec-kit phase execution without losing project context?

Automate spec-kit phase execution by loading resolved phase instructions and delegating each step sequentially to specialized sub-agents. This preserves project-root context and tracks generated files throughout the workflow.

What is the best way to orchestrate sub-agents for specification and planning workflows?

Orchestrate sub-agents for specification and planning by parsing a phase instruction file and executing one step per sub-agent in order. This sequential delegation ensures discrete task completion without mixing steps.

How do I track files created during spec-kit task execution?

Track files created during spec-kit task execution by monitoring artifacts generated by each sub-agent. The orchestrator reads these completed artifacts back to produce a content-rich final summary of the phase.

Can I execute multiple spec-kit phases simultaneously to speed up tasking workflows?

Spec-kit phase execution coordinates a single phase at a time by delegating steps sequentially to specialized sub-agents. This strict one-step-per-sub-agent approach prevents context loss and step mixing during tasking workflows.

Do I need pre-configured sub-agents to run spec-kit planning phases?

Running spec-kit planning phases requires specialized sub-agents to handle delegated steps. The orchestrator parses the phase instruction file and assigns each discrete step to a sub-agent for sequential execution.

Why does my spec-kit phase summary not reflect the completed artifacts accurately?

Spec-kit phase summaries are generated from the exact contents of completed artifacts read back after execution. If sub-agents do not finish steps sequentially or file tracking misses artifacts, the final summary will lack accurate content.