autoresearch

autoresearch is a skill for Claude Code, Codex from PaulRBerg/agent-skills. It costs 32 tokens per session (948 once invoked), scanned A, original, MIT.

A workflow for running repeated, measured experiments to improve something over time. It plans bounded batches, sends implementation work to separate workers, and keeps only verified improvements.

In plain words
What is it for?
Use it to test competing changes overnight or in a loop, compare results against a metric, retain changes that improve the result, and stop when resources or progress run out.
Why use it?
It gives open-ended optimization work a measurable goal, limits, validation, and a stopping point.

Skill for Claude CodeCodex

Written for Claude Code and Codex: argument-hint in frontmatter, but also agents/openai.yaml present. Also seen: mentions Codex; $skill-name invocation.

Good fit Use it to test competing changes overnight or in a loop, compare results against a metric, retain changes that improve the result, and stop when resources or progress run out.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/paulrberg/agent-skills/autoresearch
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.

Any agent
npx skills add PaulRBerg/agent-skills --skill autoresearch
Clone the repo
git clone --depth 1 https://github.com/PaulRBerg/agent-skills

Made for: Claude Code, 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

README.md
[![agentmods](https://agentmods.dev/badge/skills/paulrberg/agent-skills/autoresearch/github.svg)](https://agentmods.dev/skills/paulrberg/agent-skills/autoresearch)
Your own site
<a href="https://agentmods.dev/skills/paulrberg/agent-skills/autoresearch"><img src="https://agentmods.dev/badge/skills/paulrberg/agent-skills/autoresearch/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

Your own site · 80×15
<a href="https://agentmods.dev/skills/paulrberg/agent-skills/autoresearch"><img src="https://agentmods.dev/badge/skills/paulrberg/agent-skills/autoresearch.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 32 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 948 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits
  • NVIDIA SkillSpector pass 7 Sept 2026
How audits are shown
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 $0.00032 $0.00948
Opus 5 $0.00016 $0.00474
Sonnet 5 $0.00006 $0.00190
Haiku 4.5 $0.00003 $0.00095

Measured 13d ago against content hash 147d31d84d92, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-12, from the pricing page.

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 13d ago.

The scan reads SKILL.md. This mod also ships 3 executable files (scripts/autoresearch-session.py, scripts/confidence.sh, scripts/summary.sh), 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.

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.

skills/autoresearch/SKILL.md · 77 lines

How it starts

The opening of the file, as written. The whole thing — 77 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Autoresearch

Use the parent for research decisions and Codex handoff workers for experiment execution. Measure consistently, retain only verified improvements, and stop on explicit resource or convergence limits.

Orchestration Contract

Start a new autoresearch session in Plan mode. Also require Plan mode before materially changing an approved session's objective, primary metric or direction, benchmark or correctness commands, write scope, or hard resource, cost, regression, or convergence limits. If Plan mode is required but inactive, ask the user to switch and stop. An unchanged approved contract may resume and receive new hypothesis batches outside Plan mode.

Always invoke $codex-handoff for implementation. Never fall back to direct parent implementation or another handoff mechanism. Follow its host selection, plan manifest, team sizing, validation ownership, reconciliation, failure, and completion contracts.

The parent owns the session contract, evidence synthesis, hypothesis selection and ordering, and stop decisions. It may delegate read-only repository investigation when useful, but research workers return evidence rather than hypotheses or plans. Keep the parent's execution work to orchestration, integrity checks, compact result review, and selection of the next batch.

Implementation workers execute parent-supplied ordered hypothesis batches. Default to one worker for a sequential search and pack multiple related hypotheses into one brief up to codex-handoff's sizing limit; never map one worker to each idea by default. Split only when hypotheses are genuinely independent, dependency waves require it, or one brief would be oversized. A worker may make local adjustments needed to execute an assigned hypothesis, but it must report rather than execute a materially different research direction.

Plan and Session Contract

Resolve the objective, primary metric and direction, benchmark and correctness commands, allowed/off-limits paths, run/runtime/command/cost/regression limits, convergence window, and reporting cadence before approval. Infer safe facts from the request and repository; ask only when a missing choice changes the experiment.

Read the full file on GitHub · 77 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. 13d ago First seen · 77 lines · 32 tokens per session scan A 147d31d84d92

Subscribe to this mod's changes

autoresearch is a skill published in the GitHub repository PaulRBerg/agent-skills (70 stars, last pushed 2d ago), licensed MIT. It adds 32 tokens to every session and 948 once invoked, about $0.0002 per session on Opus 5. 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.

Related

Other skills, from other repositories

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

obra/superpowers · 21 tokens

local-ai-agents

Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…

microsoft/ai-agents-for-beginners · 200 tokens

next-cache-components-adoption

Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…

vercel/next.js · 95 tokens

chat-pet-sprite-creation

Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.

microsoft/vscode · 53 tokens

cpu-profile-analysis

Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…

microsoft/vscode · 71 tokens

insight-error-page

Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…

vercel/next.js · 83 tokens