autoresearch

Replace deprecated omx autoresearch with a stateful native-hook research loop.

Updated Apr 23, 2026

Search Tags:#automation #workflow #validation #artifact #stateful #autoresearch #native-hook

Ready to run

npx skills add https://github.com/wmasfoe/notch-pet --skill autoresearch-wmasfoe

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: autoresearch
Source: https://github.com/wmasfoe/notch-pet/tree/main/.codex/skills/autoresearch
Command: npx skills add https://github.com/wmasfoe/notch-pet --skill autoresearch-wmasfoe

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Autoresearch previously relied on a deprecated workflow; this skill provides a stateful, native-hook research loop that continuously nudges the research until validation is achieved, improving reliability and reproducibility.

Core Features & Use Cases

Init-time validation mode selection: choose between mission-validator-script or prompt-architect-artifact.
Persistent state: stores mode, artifacts, and validator data at .omx/state/.../autoresearch-state.json.
Artifact-gated completion: halts only when a valid completion artifact is produced.
Migration-friendly: removes direct CLI launches and tmux launch surfaces in favor of an integrated workflow.

Quick Start

Run the workflow and choose the validation mode at initialization to start the loop.

Frequently Asked Questions about autoresearch

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I automate a stateful research loop with artifact-based validation?▼

You can automate a stateful research loop by initializing a native-hook workflow that continuously nudges the process until a valid completion artifact is produced. This replaces deprecated CLI commands by persisting validation mode and artifact data to a JSON state file.

What is artifact-gated completion in an automated research workflow?▼

Artifact-gated completion is a validation mechanism that continuously runs a research loop and only halts when a specific, valid completion artifact is generated. This ensures the workflow meets defined validation criteria before stopping.

How do I migrate from a deprecated CLI research command to a native-hook workflow?▼

To migrate from a deprecated CLI research command, initialize a native-hook loop that removes direct CLI launches and tmux surfaces. The workflow is then guided through integrated commands and persists its state to a JSON file for reliability.

When do I need to choose between mission-validator-script and prompt-architect-artifact validation modes?▼

You need to choose between these two validation modes at initialization time when setting up an automated research loop. The selected mode determines how the workflow validates completion artifacts and persists its state throughout the process.

Does the stateful research workflow support direct CLI launches and tmux surfaces?▼

No, the stateful research workflow explicitly removes direct CLI launches and tmux surfaces. It is designed to be migration-friendly, guiding users to run the workflow via integrated commands instead of direct terminal interfaces.

Where does a stateful research loop store its persistent mode state and artifacts?▼

A stateful research loop stores its persistent mode state, validator data, and related artifacts in a JSON file at .omx/state/.../autoresearch-state.json, ensuring reproducibility across multiple workflow iterations.