Agent Plugins Marketplace
All plugins

claude-code-harness

v1.23.4

Designs, bootstraps, audits, and extends a Claude Code harness — the .claude/ directory and root CLAUDE.md. Opinion: under a capable model, less harness yields more productivity.

Claude Code1 Skill

By nikitaCodeSaveLicense: MIT5 GitHub starsUpdated 2 hours ago

Directory evidence

Runtimes
Claude Code
Parsed components
1 skill or MCP entry
Source updated
Aug 29, 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 plugin

Installs for the current user
claude plugin marketplace add IchenDEV/agent-plugin-mkt
claude plugin marketplace update agent-plugin-marketplace
claude plugin install claude-code-harness@agent-plugin-marketplace

Paste 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/nikitaCodeSave/claude-code-harness

Clone the source repository, then follow its setup instructions to add the plugin to a compatible client. The plugin root is plugins/harness/.

Plugin files

plugins/harness/
├── .claude-plugin/plugin.json
└── SKILL.md

Included Skills1

claude-code-harnessSKILL.md

Use this skill when introducing, auditing, refactoring, or extending the Claude Code harness of a project — i.e. its `.claude/` directory (CLAUDE.md, settings.json, hooks/, agents/, skills/, commands/) and root CLAUDE.md. Activates on phrases like 'set up Claude Code in this project', 'design my .claude/', 'audit my Claude Code harness', 'add a hook/skill/subagent', 'how should I organize CLAUDE.md', 'what built-ins does Claude Code already have', 'should I write a custom orchestrator subagent', 'when should I use a dynamic workflow', 'why is my harness slow / brittle / token-heavy', 'extend Claude Code with X'. Claude Code 2.x / Opus-class-specific — does NOT teach harness design for other agent frameworks (building agents in OpenAI/Codex/LangChain-style stacks); an external CLI wired into *this* harness as an MCP server is in scope. Skip when the project's `.claude/` already encodes this discipline.

Plugin manifests1

plugins/harness/.claude-plugin/plugin.json
{
  "$schema": "https://json.schemastore.org/claude-code-plugin-manifest.json",
  "name": "claude-code-harness",
  "version": "1.23.4",
  "description": "Designs, bootstraps, audits, and extends a Claude Code harness — the .claude/ directory and root CLAUDE.md. Opinion: under a capable model, less harness yields more productivity.",
  "author": {
    "name": "nikitaCodeSave",
    "email": "[email protected]"
  },
  "homepage": "https://github.com/nikitaCodeSave/claude-code-harness",
  "repository": "https://github.com/nikitaCodeSave/claude-code-harness",
  "license": "MIT",
  "keywords": [
    "claude-code",
    "harness",
    "plugin-development",
    "skills",
    "subagents",
    "bootstrap",
    "audit"
  ]
}

If you maintain this plugin, link to this source-backed listing from your README so users can review its manifest and indexed components.

[claude-code-harness on Agent Plugins Marketplace](https://pluginsmp.com/plugins/claude-code-harness)