autoresearch-agent

autoresearch-agent is a skill for Codex from bestagentkits/agency-skills. It costs 115 tokens per session (2,904 once invoked), scanned A, original, MIT.

An automated experiment loop that edits one file, measures the result with a fixed evaluation, and keeps or discards each change. It repeats this process until stopped.

In plain words
What is it for?
Use it to optimize a file for a defined measure, such as speed, size, or writing quality, and to run repeated experiments over time.
Why use it?
It removes the need to make and compare optimization changes by hand while preserving only measured improvements.

Skill for Codex

Written for Codex: agents/openai.yaml present. Also seen: reads .claude/ paths; mentions Claude Code; mentions Codex.

Needs its repository: it runs a file that does not travel with it, so clone the repository first. The line is ./scripts/convert.sh --skill autoresearch-agent --tool codex|gemini|cursor|windsurf|openclaw.

Good fit Use it to optimize a file for a defined measure, such as speed, size, or writing quality, and to run repeated experiments over time.

Compare 6 skills from other repositories ↓
Install

Getting it into your agent

It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.

Clone the repo
git clone --depth 1 https://github.com/bestagentkits/agency-skills
agentmods
npx agentmods add skills/bestagentkits/agency-skills/autoresearch-agent

Made for: Codex.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for autoresearch-agent

README.md
[![agentmods](https://agentmods.dev/badge/skills/bestagentkits/agency-skills/autoresearch-agent/github.svg)](https://agentmods.dev/skills/bestagentkits/agency-skills/autoresearch-agent)
Your own site
<a href="https://agentmods.dev/skills/bestagentkits/agency-skills/autoresearch-agent"><img src="https://agentmods.dev/badge/skills/bestagentkits/agency-skills/autoresearch-agent/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for autoresearch-agent

Your own site · 80×15
<a href="https://agentmods.dev/skills/bestagentkits/agency-skills/autoresearch-agent"><img src="https://agentmods.dev/badge/skills/bestagentkits/agency-skills/autoresearch-agent.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 115 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,904 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 1 finding. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00115 $0.02904
Opus 5 $0.00057 $0.01452
Sonnet 5 $0.00023 $0.00581
Haiku 4.5 $0.00012 $0.00290

Measured 12d ago against content hash 0c26d36d71fc, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-11, from the pricing page.

Security

Grade A, and why

autoresearch-agent scanned grade A with 1 finding 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 12d ago.

The scan reads SKILL.md. This mod also ships 3 executable files (scripts/log_results.py, scripts/run_experiment.py, scripts/setup_experiment.py), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

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.

Runs shell commandslowCapability

Expected in a hook, worth knowing in a rule or an instructions file.

result = subprocess.run(["my-benchmark", "--json"], capture_output=True, text=True)
skills/claude-skills/autoresearch-agent/SKILL.md · 309 lines

How it starts

The opening of the file, as written. The whole thing — 309 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Autoresearch Agent

You sleep. The agent experiments. You wake up to results.

Autonomous experiment loop inspired by Karpathy's autoresearch. The agent edits one file, runs a fixed evaluation, keeps improvements, discards failures, and loops indefinitely.

Not one guess — fifty measured attempts, compounding.


Slash Commands

Command What it does
/ar:setup Set up a new experiment interactively
/ar:run Run a single experiment iteration
/ar:loop Start autonomous loop with configurable interval (10m, 1h, daily, weekly, monthly)
/ar:status Show dashboard and results
/ar:resume Resume a paused experiment

When This Skill Activates

Recognize these patterns from the user:

  • "Make this faster / smaller / better"
  • "Optimize [file] for [metric]"
  • "Improve my [headlines / copy / prompts]"
  • "Run experiments overnight"
  • "I want to get [metric] from X to Y"
  • Any request involving: optimize, benchmark, improve, experiment loop, autoresearch

If the user describes a target file + a way to measure success → this skill applies.


Setup

First Time — Create the Experiment

Run the setup script. The user decides where experiments live:

Project-level (inside repo, git-tracked, shareable with team):

python scripts/setup_experiment.py \
  --domain engineering \
  --name api-speed \
  --target src/api/search.py \
  --eval "pytest bench.py --tb=no -q" \
  --metric p50_ms \
  --direction lower \
  --scope project

User-level (personal, in ~/.autoresearch/):

python scripts/setup_experiment.py \
  --domain marketing \
  --name medium-ctr \
  --target content/titles.md \
  --eval "python evaluate.py" \
  --metric ctr_score \
  --direction higher \
  --evaluator llm_judge_content \
  --scope user

The --scope flag determines where .autoresearch/ lives:

  • project (default) → .autoresearch/ in the repo root. Experiment definitions are git-tracked. Results are gitignored.
  • user~/.autoresearch/ in the home directory. Everything is personal.

Read the full file on GitHub · 309 lines

Files

What ships with it

6 files beside SKILL.md 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.

  1. 12d ago First seen · 309 lines · 115 tokens per session scan A 0c26d36d71fc

Subscribe to this mod's changes

autoresearch-agent is a skill published in the GitHub repository bestagentkits/agency-skills (11 stars, last pushed 2mo ago), licensed MIT. It adds 115 tokens to every session and 2,904 once invoked, about $0.0006 per session on Opus 5. A static security scan graded it A with 1 finding (runs shell commands). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Related

Other skills, from other repositories

exposure-risk-quantification

FAIR-aligned exposure quantification: turns a pile of recon findings into a defensible 0-100 + A-F org risk score (Likelihood x Impact, three ownership-aware factors: exposure/threat/impact), an ownership + proof demotion cap so unproven or weakly-owned findings can't inflate the number, a $-denominated FAIR…

elementalsouls/Claude-OSINT · 235 tokens

osint-methodology

Comprehensive OSINT methodology for external red-team operations and authorized attack-surface assessments. Covers the 6-stage recon pipeline (seed → asset expansion → enrichment → exposure analysis → convergence → operator-armed active validation) with connector-resilience and stage-vs-gating discipline, asset-graph…

elementalsouls/Claude-OSINT · 150 tokens

bencium-impact-designer

Create distinctive, production-grade frontend interfaces with high design quality. Use this skill when the user asks to build web components, pages, or applications. Generates creative, polished code that avoids generic AI aesthetics. Based on Anthropic's Frontend Designer Skill.

bencium/bencium-marketplace · 56 tokens

file-processing

Process and analyze CSV, JSON, and text files with data transformation, cleaning, analysis, and visualization capabilities.

aws-samples/sample-strands-agents-agentskills · 24 tokens

design-audit

Premium UI/UX design audit and refinement skill. Conducts systematic visual audits of existing apps and produces phased, implementation-ready design plans. Use this skill whenever the user asks to audit a UI, improve an app's visual design, make an interface feel more polished or premium, review design consistency…

bencium/bencium-marketplace · 140 tokens

bencium-code-conventions

Bence's code style, tech stack, and workflow conventions.

bencium/bencium-marketplace · 19 tokens