What is run-autoresearch?

A command that manages an autonomous experiment loop through an Autoresearch Orchestrator agent. It can start, resume, inspect, or cancel a research session.

Use it to run experiments, check session status, resume an active session, or cancel one while preserving its logs and kept commits.

It keeps the session state and results in project files so work can continue or be reviewed without launching duplicate sessions.

run-autoresearch — Command for Claude Code by proyecto26

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.

Clone the repo

git clone --depth 1 https://github.com/proyecto26/autoresearch-ai-plugin

Made for: Claude Code.

Or install autoresearch-ai-plugin, the plugin that ships this one along with the rest of its 2 skills, 1 command, 1 agent, 1 hook.

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.

README.md

[![agentmods](https://agentmods.dev/badge/commands/proyecto26/autoresearch-ai-plugin/run-autoresearch/github.svg)](https://agentmods.dev/commands/proyecto26/autoresearch-ai-plugin/run-autoresearch)

Your own site

<a href="https://agentmods.dev/commands/proyecto26/autoresearch-ai-plugin/run-autoresearch"><img src="https://agentmods.dev/badge/commands/proyecto26/autoresearch-ai-plugin/run-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.

Your own site · 80×15

<a href="https://agentmods.dev/commands/proyecto26/autoresearch-ai-plugin/run-autoresearch"><img src="https://agentmods.dev/badge/commands/proyecto26/autoresearch-ai-plugin/run-autoresearch.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 14 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.

When invoked 1,052 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.

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.

Model	Per session	Once invoked
Fable 5.1	$0.00014	$0.01052
Opus 5	$0.00007	$0.00526
Sonnet 5	$0.00003	$0.00210
Haiku 4.5	$0.00001	$0.00105

Measured 10d ago against content hash eaa2ef1eb930, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

run-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 10d 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.

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.

commands/run-autoresearch.md · 44 lines

How it starts

The opening of the file, as written. The whole thing — 44 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/run-autoresearch — managed experiment loop

Manage an autoresearch session for this project. Arguments: "$ARGUMENTS"

Route the request

If the argument is status (or asks how the session is going): follow the Checking Session Status procedure from the autoresearch skill — read autoresearch.jsonl and autoresearch.md, compute totals/baseline/best/confidence, and display the summary. Do not launch the agent.

If the argument is cancel or stop: follow the Cancelling an Autoresearch Session procedure from the autoresearch skill — summarize results, preserve autoresearch.jsonl/autoresearch.md and kept commits. Do not launch the agent.

Otherwise, run a session:

Detect session type.
- Existing session: autoresearch.jsonl present (the authoritative state file). Check the last {"type":"status",...} record: if its state is running (or there is no status record — legacy log) → resume, no questions needed. If it is cancelled/done/wall → the session is concluded; do not silently resume — tell the user its final state and ask whether to restart (a new segment) or leave it. Also resume-guard on goal: if the requested goal clearly differs from the log's config-header goal, ask rather than resume. If only autoresearch.md exists (interrupted setup), treat it as a new session — setup reuses the doc's parameters.
- ML training session: the goal mentions LLM/GPU training, val_bpb, pretraining — or the project contains the train.py/prepare.py template → use the autoresearch-ml skill protocol.
- Anything else → the generic autoresearch skill protocol.
Gather setup parameters (new sessions only). Required before dispatch: goal, benchmark command, primary metric (name, unit, direction), files in scope, constraints. Take them from "$ARGUMENTS" and the project; ask the user only for what cannot be inferred. Check .claude/autoresearch-ai-plugin.local.md for max_iterations, working_dir, and timeouts.

Read the full file on GitHub · 44 lines

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.

10d ago First seen · 44 lines · 14 tokens per session scan A eaa2ef1eb930

Subscribe to this mod's changes

run-autoresearch is a command published in the GitHub repository proyecto26/autoresearch-ai-plugin (12 stars, last pushed 1mo ago), licensed MIT. It adds 14 tokens to every session and 1,052 once invoked, about $0.0001 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.

Autonomous Agents Claude Code Research ai-agent autoresearch benchmarking claude-ai

run-autoresearch

Getting it into your agent

Wrote this? Show the measurements

What it costs to keep this loaded

Grade A, and why

Nothing flagged

How it starts

/run-autoresearch — managed experiment loop

Route the request

What this file has done since we first saw it

Other commands, from other repositories

discover

optimize

deep-search

git-release-workflow

review

types