reach
v0.13.0A process for building with coding agents: work split by who verifies it, documents that are capped so they get deleted from, and guards that have all been watched to fail.
By Amelia BleekerLicense: MIT0 GitHub starsUpdated 1 hour ago
Directory evidence
- Runtimes
- Claude Code
- Parsed components
- 5 skill or MCP entries
- Source updated
- Oct 2, 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 reach for Claude Code
claude plugin marketplace add IchenDEV/agent-plugin-mkt
claude plugin marketplace update agent-plugin-marketplace
claude plugin install reach@agent-plugin-marketplacePaste and run these commands in a terminal with Claude Code. They add and refresh the PluginsMP catalog, then install this plugin.
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/SpoiledInk13/ReachClone the source repository, then follow its setup instructions to add the plugin to a compatible client. The repository root is the plugin root.
Plugin files
├── .claude-plugin/plugin.json├── skills/adopt/SKILL.md├── skills/build/SKILL.md├── skills/ideate/SKILL.md├── skills/iterate/SKILL.md└── skills/milestone/SKILL.md
Included Skills5
Replace a repository's existing agent workflow with this one, by distilling what is already there rather than adding beside it. Use when a project is adopting reach for the first time, or when its CLAUDE.md and commands have stopped being read.
Implement and prove everything that is not presentation, reading the unit documents as the backlog. Runs until nothing is left that can be built without an answer. Use to build unbuilt units, satisfy owed claims, or work through the roster.
The conversation that owns the documents. Talk an idea through, then write it down — replacing rather than appending. The only command that answers a question, decides a design, or changes what a document says. Use for design conversations, architecture decisions, answering Open questions filed by other commands, and tooling.
Run a lane unattended from this window — check nothing holds it, start its supervisor, follow what each run lands, and stop it on request. This window never builds. Use to run the builder non-stop, to see how an unattended lane is getting on, or to stop one.
Build the layer only a person can verify — how the software looks, sounds and feels — one walkthrough at a time, hand the owner a real build, and delete the walkthrough when they confirm it. Use for UI, visual and audio work whose verdict is the owner's eyes and ears.
Plugin manifests1
{
"name": "reach",
"displayName": "Reach",
"version": "0.13.0",
"description": "A process for building with coding agents: work split by who verifies it, documents that are capped so they get deleted from, and guards that have all been watched to fail.",
"author": {
"name": "Amelia Bleeker"
},
"license": "MIT",
"skills": "./skills/"
}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.
[reach on Agent Plugins Marketplace](https://pluginsmp.com/plugins/reach)