autoresearch

An automated experiment loop that tries successive changes to reach a plain-language goal. It chooses how to create and judge the result, then can compare multiple attempts when progress slows.

In plain words
What is it for?
Use it to improve an artifact through repeated experiments, such as optimizing a result against a number score, a pass-or-fail condition, or written review criteria.
Why use it?
It reduces the need to manually design every experiment, score each result, and decide what to try next.

Plugin for Claude Code

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.

Claude Code
/plugin marketplace add FradSer/dotclaude
agentmods
npx agentmods add plugins/fradser/dotclaude/autoresearch
Clone the repo
git clone --depth 1 https://github.com/FradSer/dotclaude

Made for: Claude Code.

Per session not measured What this adds to a session before it is invoked.
When invoked not measured Not applicable: nothing here is loaded into a session.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Security

Grade A, and why

autoresearch 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 yesterday.

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.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

autoresearch/.claude-plugin/plugin.json · 36 lines

What it actually says

{
  "name": "autoresearch",
  "version": "0.9.2",
  "description": "Autonomous research loop inspired by karpathy/autoresearch — give it a plain-language goal and it infers the artifact, evaluator (numeric scorer / objective gate / LLM rubric), and bounds, then iterates: cheap sequential rounds that escalate to a parallel GAN tournament when they plateau",
  "author": {
    "name": "Frad LEE",
    "email": "[email protected]"
  },
  "license": "MIT",
  "keywords": [
    "research",
    "autonomous",
    "loop",
    "experiment",
    "optimization",
    "scorer",
    "evaluator",
    "gan",
    "tournament",
    "workflow",
    "rubric"
  ],
  "hooks": {
    "Stop": [
      {
        "hooks": [
          {
            "type": "command",
            "command": "${CLAUDE_PLUGIN_ROOT}/hooks/stop-hook.sh"
          }
        ]
      }
    ]
  }
}
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.

  1. yesterday First seen · 36 lines scan A 345cc5f03d66

Subscribe to this mod's changes

autoresearch is a plugin published in the GitHub repository FradSer/dotclaude (587 stars, last pushed 20d ago), licensed MIT. Its token cost is not measured: this kind of file is read by the harness, not the model. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.