loop-graph-design
v0.9.0Interview the owner into a grounded agent loop, then emit the runnable prompt or script that starts it.
By rocky24310 GitHub starsUpdated 45 minutes ago
Directory evidence
- Runtimes
- Codex
- Parsed components
- 1 skill or MCP entry
- Source updated
- Sep 3, 2026
- Manifest status
- Canonical path parsed
The directory validates manifest shape and source location. It does not execute the plugin or provide a security endorsement. Review the indexing methodology →
Install loop-graph-design for Codex
codex plugin marketplace add rocky2431/loop-graph-design-skill
codex plugin marketplace upgrade rocky-loop-graph-design
codex plugin add loop-graph-design@rocky-loop-graph-designPaste and run these commands in a terminal with Codex. They add and refresh the rocky-loop-graph-design catalog, then install this plugin.
Compatibility: the page URL and API slug “loop-graph-design” remain stable.
- Codex:
loop-graph-design@agent-plugin-marketplace→loop-graph-design@rocky-loop-graph-design
The installer fetches third-party code from the source repository shown on this page. This directory validates manifest structure and source location, but does not perform a security audit; review the manifest, components, and source before installing.
Get the source manually
git clone https://github.com/rocky2431/loop-graph-design-skillClone the source repository, then follow its setup instructions to add the plugin to a compatible client. The plugin root is plugins/loop-graph-design/.
Plugin files
├── .codex-plugin/plugin.json└── skills/loop-graph-design/SKILL.md
Included Skills1
Turn "make an agent keep doing this" into a goal the host will actually hold to: interview for intent, anchor, quantified stop condition, boundary, and an independent verifier, refuse the shapes that fail, then emit the runnable artifact — a goal line to paste, a workflow script, or a cross-vendor delegation package. Use when the deliverable is runnable, not a design note.
Plugin manifests1
{
"name": "loop-graph-design",
"version": "0.9.0",
"description": "Interview the owner into a grounded agent loop, then emit the runnable prompt or script that starts it.",
"author": {
"name": "rocky2431",
"url": "https://github.com/rocky2431"
},
"repository": "https://github.com/rocky2431/loop-graph-design-skill",
"skills": "./skills/",
"interface": {
"displayName": "Loop Graph Design",
"shortDescription": "Turn repeating work into a grounded running loop",
"longDescription": "Adds a provider-neutral Skill that interviews the owner for a loop's intent, anchor, quantified stop condition, boundary, and independent verifier, refuses the topologies that fail in production, and compiles a machine-consumable artifact: a /goal or /loop prompt package, a Workflow script, or a cross-vendor delegation package. Topology stays model-authored; only the package surface is mechanically checked.",
"developerName": "rocky2431",
"category": "Developer Tools",
"capabilities": [
"Skills"
],
"defaultPrompt": "Use $loop-graph-design to turn repeating work into a loop with an anchor and an owner-defined stop condition."
}
}For maintainers
If you maintain this plugin, link to this source-backed listing from your README so users can review its manifest and indexed components.
[loop-graph-design on Agent Plugins Marketplace](https://pluginsmp.com/plugins/loop-graph-design)