autoresearch

A system for repeatedly testing changes against a chosen goal and measurement. It keeps changes that improve the result, reverses changes that do not, and records each run.

In plain words
What is it for?
Use it to optimize a measurable result in a code repository, such as a benchmark or model score.
Why use it?
It removes the need to manage every optimization attempt manually while preserving a history of what was tried.

Skill for Claude CodeCodex

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.

agentmods
npx agentmods add skills/baleen37/bstack/autoresearch
Any agent
npx skills add baleen37/bstack --skill autoresearch
Clone the repo
git clone --depth 1 https://github.com/baleen37/bstack

Made for: Claude Code, Codex.

Per session 46 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,443 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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 $0.00046 $0.02443
Opus 5 $0.00023 $0.01222
Sonnet 5 $0.00009 $0.00489
Haiku 4.5 $0.00005 $0.00244

Measured 2d ago against content hash 8112f4cbc392, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, 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 2d ago.

The scan reads SKILL.md. This mod also ships 1 executable file (scripts/loop.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.

plugins/autoresearch/skills/autoresearch/SKILL.md · 222 lines

How it starts

The opening of the file, as written. The whole thing — 222 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Autoresearch

Autonomous experiment loop: try ideas, keep what works, revert what doesn't, and record every run.

Modeled on karpathy/autoresearch, generalized from "optimize val_bpb in train.py" to any metric in any repo.

Setup

  1. Ask (or infer): Goal, Command, Metric (+ direction), Files in scope, Constraints.
  2. Check git status and the current branch. Preserve unrelated user changes, then create a fresh, unique branch such as autoresearch/<goal>-<date>; do not reuse an existing branch.
  3. Read the source files. Understand the workload deeply before writing anything.
  4. mkdir -p .autoresearch, then write .autoresearch/autoresearch.md and .autoresearch/run.sh. If looping was explicitly requested, copy the bundled scripts/loop.sh resource to .autoresearch/loop.sh and make it executable. Commit only those setup files. Keep .autoresearch/results.jsonl untracked.
  5. Run the unchanged baseline first. Append it as the first results.jsonl line, validate the JSONL, and only then start experimenting.

The default operation is exactly one iteration: setup or resume, run one benchmark, keep or revert one change, append one ledger row, then stop. Repeat only when the user explicitly asks for a loop or starts .autoresearch/loop.sh.

autoresearch.md

This is the heart of the session — upstream's program.md. A fresh agent with no context should be able to read this file and run the loop effectively. Invest time making it excellent.

# Autoresearch: <goal>

## Objective
<Specific description of what we're optimizing and the workload.>

## Metrics
- **Primary**: <name> (<unit>, lower/higher is better)
- **Secondary**: <name>, <name>, ...

## How to Run
`./.autoresearch/run.sh` — outputs `METRIC name=number` lines.

## Files in Scope
<Every file the agent may modify, with a brief note on what it does.>

## Off Limits
<What must NOT be touched.>

## Constraints
<Hard rules: tests must pass, no new deps, etc.>

## What's Been Tried
<Update as experiments accumulate. Note key wins, dead ends, and architectural
insights so the agent doesn't repeat failed approaches.>

Read the full file on GitHub · 222 lines

Files

What ships with it

1 file 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. 2d ago First seen · 222 lines · 46 tokens per session scan A 8112f4cbc392

Subscribe to this mod's changes

autoresearch is a skill published in the GitHub repository baleen37/bstack (4 stars, last pushed 12d ago), licensed MIT. It adds 46 tokens to every session and 2,443 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 2026-08-31.

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

brainstorming

You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.

obra/superpowers · 37 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

agent-host-chat-contributions

Build and review cross-cutting agent-host chat behavior through lifecycle contributions. Use when adding turn lifecycle side effects, prompt or context injection, restored-history transformation, protocol-action observation, or when reviewing changes that add code to AgentSideEffects or AgentService.

microsoft/vscode · 56 tokens

auto-perf-optimize

Run agent-driven VS Code performance or memory investigations. Use when asked to launch Code OSS, automate a VS Code scenario, run the Chat memory smoke runner, capture renderer heap snapshots, take workflow screenshots, compare run summaries, or drive a repeatable scenario before heap-snapshot analysis.

microsoft/vscode · 62 tokens