autoimprove

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Security

Grade A, and why

autoimprove scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

.claude-plugin/marketplace.json · 27 lines

What it actually says

{
  "name": "autoimprove",
  "owner": {
    "name": "Ben Marte",
    "email": "[email protected]"
  },
  "metadata": {
    "description": "Autonomous codebase improvement loop for Claude Code using isolated git worktrees.",
    "version": "1.0.0"
  },
  "plugins": [
    {
      "name": "autoimprove",
      "source": "./",
      "description": "Language-agnostic autonomous codebase improvement loop using isolated git worktrees. Every experiment runs on its own branch — main is never touched until a winning change is merged.",
      "category": "development",
      "tags": [
        "autoimprove",
        "code-quality",
        "autonomous",
        "worktrees",
        "improvement-loop"
      ]
    }
  ]
}

Files

What ships with it

2 files beside marketplace.json in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

Related

Other plugins, from other repositories