skills.rest
prthik
prthikCOMMUNITY
Explore all skills by this author →
GitHub

autoresearch

Run stateful research loops gated by validator scripts or prompts.

1|Updated Sep 22, 2025
Search Tags:#automation#research#validation#artifact#stateful#ai-workflow#autoresearch
One-click install
npx skills add https://github.com/prthik/prathik-astro --skill autoresearch-prthik
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: autoresearch
Source: https://github.com/prthik/prathik-astro/tree/main/.codex/skills/autoresearch
Command: npx skills add https://github.com/prthik/prathik-astro --skill autoresearch-prthik

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Autoresearch replaces the deprecated omx autoresearch command by providing a stateful, validator-gated research loop that persists via a native-hook mechanism, enabling continued evaluation across runs without manual re-launch.

Core Features & Use Cases

  • Init chooses exactly one validation mode at startup: mission-validator-script or prompt-architect-artifact.
  • Persist mode state in .omx/state/.../autoresearch-state.json including validation_mode, completion_artifact_path, mission_validator_command or validator_prompt, and optional output_artifact_path.
  • Completion is artifact-gated; the loop continues until a valid completion artifact exists and meets the chosen validation regime.
  • Direct CLI launch is gone; use $deep-interview --autoresearch for intake and $autoresearch for execution.

Quick Start

Initiate the workflow by selecting a validation mode, then run the autoresearch loop to produce a gated completion artifact.

Frequently Asked Questions about autoresearch

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

FAQPage Schema
How do I automate long-running research loops with persistent state across runs?▼

You automate long-running research loops by using a stateful, validator-gated mechanism that persists progress in a JSON state file, enabling evaluation to continue across multiple runs without manual re-launch.

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

Artifact-gated completion means the research loop continues iterating until a valid completion artifact is produced and successfully meets the validation regime specified at initialization.

How do I set up validator-gated research using a mission validator command?▼

You set up validator-gated research by initiating the workflow via the deep-interview command to select a validation mode, which configures the mission_validator_command or validator_prompt in the persistent state file.

Can I use prompt-architect-artifact validation for iterative AI workflows?▼

Yes, prompt-architect-artifact is one of the validation modes selectable at startup to gate the completion of iterative AI research workflows using a specific prompt architecture.

Does autoresearch support direct CLI launch for workflow execution?▼

No, direct CLI launch is deprecated; workflow execution requires using the deep-interview command for intake and the autoresearch command for running the stateful research loop.

Related Skills

Explore More Education & Research Skills→
wmasfoewmasfoe

autoresearch

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

Community
Advanced
fagemxfagemx

wusanto

Enforce verification gates and mission-based contracts for AI-generated work.

Community
Advanced
OrinksOrinks
24

autoresearch

Persist validator-gated research loops with init-time mode selection.

Community
Advanced
wanshuiyinwanshuiyin
14.4k

auto-review-loop-minimax

Automate multi-round research review loops using the MiniMax API.

Community
Advanced