autoresearch-orchestrator

autoresearch-orchestrator is an agent for Claude Code from proyecto26/autoresearch-ai-plugin. It costs 115 tokens per session (1,884 once invoked), scanned A, original, MIT.

An agent that manages an autoresearch session from setup and baseline measurement through repeated code experiments.

In plain words
What is it for?
Use it to start or resume an optimization session with a goal, benchmark command, metric, files, and constraints. It handles edit, measure, keep-or-revert cycles and returns a checkpoint.
Why use it?
It keeps the experiment process organized, including recording state, comparing measurements, and continuing a session after an interruption.

Agent for Claude Code

Written for Claude Code: ${CLAUDE_SKILL_DIR} variable. Also seen: model in frontmatter.

Part of the autoresearch-ai-plugin plugin — 2 skills, 1 command, 1 agent, 1 hook shipped together

Good fit Use it to start or resume an optimization session with a goal, benchmark command, metric, files, and constraints. It handles edit, measure, keep-or-revert cycles and returns a checkpoint.

Compare 6 agents from other repositories ↓
Install with agentmods
npx agentmods add agents/proyecto26/autoresearch-ai-plugin/autoresearch-orchestrator
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.

agentmods badge for autoresearch-orchestrator

README.md
[![agentmods](https://agentmods.dev/badge/agents/proyecto26/autoresearch-ai-plugin/autoresearch-orchestrator/github.svg)](https://agentmods.dev/agents/proyecto26/autoresearch-ai-plugin/autoresearch-orchestrator)
Your own site
<a href="https://agentmods.dev/agents/proyecto26/autoresearch-ai-plugin/autoresearch-orchestrator"><img src="https://agentmods.dev/badge/agents/proyecto26/autoresearch-ai-plugin/autoresearch-orchestrator/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-orchestrator

Your own site · 80×15
<a href="https://agentmods.dev/agents/proyecto26/autoresearch-ai-plugin/autoresearch-orchestrator"><img src="https://agentmods.dev/badge/agents/proyecto26/autoresearch-ai-plugin/autoresearch-orchestrator.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 115 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,884 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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00115 $0.01884
Opus 5 $0.00057 $0.00942
Sonnet 5 $0.00023 $0.00377
Haiku 4.5 $0.00012 $0.00188

Measured 10d ago against content hash 173718dc232b, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

autoresearch-orchestrator 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.

agents/autoresearch-orchestrator.md · 61 lines

How it starts

The opening of the file, as written. The whole thing — 61 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are the Autoresearch Orchestrator: you manage an autonomous optimization loop — edit code, commit, run a benchmark, measure the primary metric, keep improvements or revert — and you report verifiable results. You follow the plugin's autoresearch skill protocol (and the autoresearch-ml specialization for GPU LLM training) exactly as written in the SKILL.md files bundled with this plugin.

When to invoke

  • New session. The /run-autoresearch command hands you a goal, benchmark command, primary metric (name, unit, direction), files in scope, and constraints. Run the full setup phase, record the baseline, then start experimenting.
  • Resume. autoresearch.jsonl exists in the working directory (the authoritative session state). Read it plus autoresearch.md (reconstruct the doc from the JSONL config header and git log if it is missing), verify git state, and continue from the last run — no re-setup, no asking for permission.
  • Continue after checkpoint. A previous orchestrator dispatch returned a checkpoint and the main conversation relaunches you to keep going. Same as resume.

Session workflow

Phase 0 — Preflight. Read .claude/autoresearch-ai-plugin.local.md if present (enabled, max_iterations, working_dir, benchmark_timeout, checks_timeout). max_iterations: 0 or absent means unlimited — never treat 0 as "already reached". If enabled: false, stop and report why. Verify git is available and the working tree state is clean or explainable. Resume rule: if autoresearch.jsonl exists, this is a resume — skip to Phase 3 (rebuilding autoresearch.md first if it is missing). If only autoresearch.md exists (interrupted setup), run Phase 1 again reusing its parameters.

Phase 1 — Setup (per the skill's Setup Phase): create the autoresearch/<goal>-<date> branch; gitignore the living session files (autoresearch.jsonl, autoresearch.md, autoresearch.ideas.md, run.log); read the files in scope; write autoresearch.md and autoresearch.sh (benchmark emitting METRIC name=value lines), optionally autoresearch.checks.sh; commit only the immutable harness (autoresearch.sh + optional autoresearch.checks.sh), never the gitignored living files.

Read the full file on GitHub · 61 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.

  1. 10d ago First seen · 61 lines · 115 tokens per session scan A 173718dc232b

Subscribe to this mod's changes

autoresearch-orchestrator is an agent published in the GitHub repository proyecto26/autoresearch-ai-plugin (12 stars, last pushed 1mo ago), licensed MIT. It adds 115 tokens to every session and 1,884 once invoked, about $0.0006 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 agents, from other repositories

autonomous-creativity-engine

Implements genuine creativity algorithms that generate novel ideas with artistic and scientific innovation systems and philosophical reasoning frameworks.

bejranonda/LLM-Autonomous-Agent-Plugin-for-Claude · 25 tokens

dimensional-computing-framework

Implements multi-dimensional data processing beyond 3D space with hyper-dimensional pattern recognition and time-space manipulation.

bejranonda/LLM-Autonomous-Agent-Plugin-for-Claude · 27 tokens

quantum-computing-integrator

Implements quantum-inspired algorithms and quantum entanglement simulation for exponential performance gains and perfect reliability.

bejranonda/LLM-Autonomous-Agent-Plugin-for-Claude · 26 tokens

ideator

Generates ranked experiment proposals for the evo orchestrator. Runs ONE brief per invocation (failureanalysis, literature, or frontierextrapolation) and appends proposals as JSONL lines to a shared file the orchestrator reconciles. Use literature for web/arXiv/HF/GitHub research (the only brief that needs network).…

evo-hq/evo · 142 tokens

architect

System architecture and design specialist. Use for designing scalable systems, making architectural decisions, creating technical specifications, or refactoring system structure.

a-ariff/ariff-claude-plugins · 28 tokens

backend-dev

Backend development specialist for APIs, databases, server-side logic, and system integration. Use for building REST/GraphQL APIs, database design, authentication, and backend architecture.

a-ariff/ariff-claude-plugins · 37 tokens