using-kilroy

Convert English requirements into Kilroy Attractor pipelines via the Kilroy CLI.

216|52|Updated Feb 6, 2026
One-click install
npx skills add https://github.com/danshapiro/kilroy --skill using-kilroy
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: using-kilroy
Source: https://github.com/danshapiro/kilroy/tree/main/skills/using-kilroy
Command: npx skills add https://github.com/danshapiro/kilroy --skill using-kilroy

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill enables operating the Kilroy CLI to turn English requirements into Attractor pipelines and orchestrate their execution inside an isolated git worktree, delivering auditable results.

Core Features & Use Cases

  • Ingest requirements to a DOT pipeline: convert natural language requirements into a Graphviz graph that Kilroy can execute.
  • Validate and run pipelines: ensure syntactic correctness, then execute node-by-node with tool-using coding agents.
  • Resume and audit runs: support resuming interrupted executions with per-node checkpoints and CXDB logging.

Quick Start

kilroy attractor ingest -o pipeline.dot "Build a Go CLI link checker with robots.txt support" kilroy attractor validate --graph pipeline.dot kilroy attractor run --graph pipeline.dot --config run.yaml

Frequently Asked Questions about using-kilroy

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

FAQPage Schema
How do I convert English requirements into an executable pipeline graph?

To validate an Attractor pipeline before execution, run the Kilroy CLI validate command against your generated Graphviz DOT file. This checks the syntactic correctness of the pipeline graph to prevent execution errors during the automated run.

Can I resume an interrupted pipeline execution with per-node checkpoints?

Yes, you can execute an Attractor pipeline using tool-using coding agents. Kilroy runs the validated Graphviz DOT graph node-by-node, orchestrating automated execution inside an isolated git worktree based on your run.yaml configuration.

Do I need a local Kilroy binary and CXDB to run pipeline automation?

Yes, you need a local Kilroy binary and a running CXDB to run pipeline automation. These components are required dependencies for ingesting requirements, executing pipelines, and logging the auditable state of node-by-node runs.

What is the best way to automate LLM pipeline ingestion from natural language requirements?

The best way to automate LLM pipeline ingestion from natural language requirements is using the Kilroy CLI. It transforms English text into a Graphviz DOT pipeline file, which can then be validated and run with tool-using coding agents in an isolated worktree.