Agent Sh Enhance
agent-sh-enhance is a Claude Code plugin for the Ship phase that analyzes plugin structure, MCP tools, and security patterns in your bundle.
by agent-sh · github.com/agent-sh/enhance
Audit Claude Code plugin bundles for valid plugin.json, MCP tool definitions, and security patterns before you ship or publish them.
Add it to Claude Code
Install the plugin in Claude Code. One command, paste-ready.
/plugin install agent-sh-enhance@agent-sh/enhanceBuilt to be called by your agent
Skillselion is itself an MCP server. Your agent can pull this entry and a paste-ready install config straight from the API - no copy-paste.
Retrieve this entry with skillselion.get_details("plugin:agent-sh/enhance") and the paste-ready config with skillselion.get_install_config("plugin:agent-sh/enhance").
What it does
agent-sh-enhance is a Claude Code plugin analyzer aimed at solo builders and small teams who assemble custom agent stacks from multiple plugins and MCP servers. Instead of guessing whether a plugin.json is complete or whether a tool surface is too permissive, you run a structured pass that ties manifest structure, MCP wiring, and security heuristics into one review. It fits the moment you are polishing a plugin marketplace entry, hardening an internal Claude Code bundle, or debugging why parallel plugin orchestration behaves oddly. The keyword set (orchestrator, parallel plugins, unified reporting) signals it is built for composite setups rather than a single skill file. Complexity is intermediate: you need basic familiarity with plugin.json and MCP concepts. Install when you already have plugins drafted and want evidence-backed fixes before ship, not when you are still ideating product scope.
Highlights
- Validates plugin.json structure and Claude Code plugin conventions
- Analyzes MCP tool definitions and tool-use patterns for risk
- Checks security patterns across orchestrated and parallel plugin setups
- Supports unified reporting across prompts, plugins, and execution paths
- Oriented to agent-sh enhance / orchestrator-style multi-plugin workflows
Why builders use it
Custom Claude Code plugins and MCP stacks are easy to misconfigure, leaving broken manifests or over-broad tools you only discover after users install them.
You get a structured validation pass on plugin.json, MCP tooling, and security patterns so you can fix issues before publishing or deploying the bundle.
At a glance
- Type - Plugin in Security.
- Adoption - 0 installs, 2 stars, 0 votes.
FAQ
Who is agent-sh-enhance for?
Solo and small-team builders who maintain Claude Code plugins, MCP servers, or multi-plugin orchestrators and need manifest and security validation before release.
When should I use agent-sh-enhance?
Use it when a plugin bundle is feature-complete but you have not yet verified plugin.json, MCP tool shapes, and security patterns—especially with parallel or unified orchestration.
How do I add agent-sh-enhance to my agent?
Install the agent-sh/enhance Claude Code plugin from the catalog, open your repo with plugin.json and MCP configs, and invoke the enhance analyzer workflow per the plugin README.
Comments
Share how you use agent-sh-enhance, gotchas, or tips for other indie builders.
No comments yet - be the first to share how you use it.