autoresearch-goal

Automate rubric-validated research planning and completion for Omagy goals.

3|Updated May 24, 2026
One-click install
npx skills add https://github.com/shayne-snap/oh-my-antigravity --skill autoresearch-goal
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: autoresearch-goal
Source: https://github.com/shayne-snap/oh-my-antigravity/tree/main/plugin/skills/autoresearch-goal
Command: npx skills add https://github.com/shayne-snap/oh-my-antigravity --skill autoresearch-goal

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It prevents one-off, ad hoc investigations by forcing research to follow an explicit professor-critic rubric and only finish when validation evidence proves the rubric is satisfied.

Core Features & Use Cases

  • Rubric-governed research mission: turns a topic plus a PASS/FAIL validation standard into a durable research plan with a control plane.
  • Professor-critic verdict tracking: records pass/fail/block decisions with evidence so decisions are auditable, not implied.
  • Durable completion gate: only marks the goal complete after a final audit and a recorded verdict:"pass" with passed:true, avoiding premature or unsupported conclusions.
  • Best-evidence finalization: separates evidence from inference and preserves the strongest final artifacts for later reuse.

Quick Start

Use Omagy autoresearch-goal to create and execute a rubric-driven research mission for the question you care about, then proceed until a critic verdict passes and completion evidence confirms the rubric.

Frequently Asked Questions about autoresearch-goal

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

FAQPage Schema
How do I run rubric-checked research with proof for an academic topic?

Rubric-checked research with proof is automated by creating a durable research plan that applies a professor-critic workflow, gathers supporting sources, and records iterative verdicts to validate completion.

How does a professor-critic workflow track research verdicts?

A professor-critic workflow tracks research verdicts by recording pass, fail, or block decisions alongside gathered evidence, ensuring that all validation outcomes are explicitly auditable rather than implied.

What prevents premature conclusions when validating research evidence?

Premature conclusions during research validation are prevented by a durable completion gate that only marks a goal complete after a final audit and a recorded verdict:"pass" with passed:true.

Can I align research goals with external host goal snapshots?

You can align research goals with external host goal snapshots by optionally accepting host goal snapshots via the --host-goal-json flag for alignment validation during the research mission.

How do I store durable artifacts for autoresearch goals?

Durable artifacts for autoresearch goals are stored under the dedicated autoresearch goals directory, using artifact-based truth to establish goal parity and preserve the strongest final evidence for later reuse.

What is the best way to separate evidence from inference in research workflows?

The best way to separate evidence from inference in research workflows is through best-evidence finalization, which isolates supporting sources from analytical conclusions and preserves the strongest final artifacts.