NORA: A Harness-Engineered Autonomous Research Agent for End-to-End Spatial Data Science

arXiv 2026 59.6 method

TLDR

NORA is a harness-engineered multi-agent system for end-to-end autonomous spatial data science with 21 domain-specialized skills and case-study evaluation.

Reasoning

The paper presents a novel domain-specific autonomous research agent with a well-defined harness engineering framework, but its evaluation relies on subjective case studies rather than quantitative benchmarks, and the abstract lacks details on reproducibility and comparison to baselines.

Read-first score

Read-first score 59.6, weighted from topical fit, citation, graph, method, reproducibility, and recency signals. Original total remains 70.

Recency 8%

100

Uses a gentle age decay so recent papers surface without erasing older foundations. 2026

Methodology quality 25%

Screens visible abstract and analysis fields for experiment, dataset, baseline, metric, and limitation evidence. markers=analysis,evaluation,result

Topical relevance 42%

58.3

Uses existing LLM keyword relevance scores normalized to 0-100. AI scientist,automated scientific discovery,autonomous research agent,automated research,literature review agent,survey generation,automated experimentation,experiment design agent,AI for scientific research,paper writing agent,research automation,scientific discovery agent

Reproducibility 25%

Screens links and visible text for paper, code, dataset, artifact, and repository signals. pdf=True; code=False; dataset=False; markers=code

Field roles

FrontierMethodology anchor

Rank sensitivity

Stability: volatile; rank range: 32.

Keyword Scores

autonomous research agent

automated research

research automation

automated scientific discovery

scientific discovery agent

AI for scientific research

AI scientist

automated experimentation

experiment design agent

literature review agent

survey generation

paper writing agent

Deep Analysis

Innovations

NORA: a harness-engineered multi-agent autonomous research agent purpose-built for spatial data science, orchestrating the complete research lifecycle.
Two novel domain-specialized skills: a spatial analysis skill unit encoding decision frameworks for exploratory spatial data analysis, spatial regression, and diagnostics; and a spatial data download skill for reproducible acquisition from authoritative geospatial sources.
Formalization of harness engineering for scientific research agents, incorporating lifecycle hooks, safety gates, generator-evaluator separation, human-in-the-loop, and state persistence.

Methodology

NORA employs a skills-first architecture with 21 domain-specialized workflow skills, 9 specialist sub-agents, and custom Model Context Protocol (MCP) servers. The system integrates harness engineering principles. Evaluation is conducted through case studies assessed by 6 domain specialists and 3 LLM reviewers across seven dimensions including novelty, quality, and rigor.

Key Results

Domain-specialized harness engineering substantially improves the efficiency and quality of research output compared to general-purpose agent configurations.

Resources

Paper