batch-analysis

Execute queued marimo notebooks headlessly and generate morning review summaries.

Updated Feb 17, 2026
One-click install
npx skills add https://github.com/etoyama/insight-blueprint --skill batch-analysis
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: batch-analysis
Source: https://github.com/etoyama/insight-blueprint/tree/main/skills/batch-analysis
Command: npx skills add https://github.com/etoyama/insight-blueprint --skill batch-analysis

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill automates unattended overnight processing of queued analysis designs by generating standardized marimo notebooks, executing them headlessly, performing an internal self-review, recording structured journal events, and producing a consolidated morning summary for human triage. It eliminates the need for manual monitoring and accelerates morning review workflows.

Core Features & Use Cases

  • Headless Batch Execution: Generate 8-cell marimo notebooks per design, execute them, and verify required outputs.
  • Automated Self-Review & Journal Recording: Apply a deterministic self-review protocol, extract evidence/observations/questions, and append structured events to design journals.
  • Robust Orchestration & Error Handling: Enforce package allowlist, marimo version preflight, a 3-attempt repair loop, and per-run lessons to handle common failures.
  • Use Case: Schedule overnight processing of multiple hypothesis-driven analyses so stakeholders receive a prioritized morning summary and notebooks ready for human reflection.

Quick Start

Run the batch-analysis skill to process all queued designs overnight and generate a morning summary for human triage.

Frequently Asked Questions about batch-analysis

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

FAQPage Schema
How do I automate overnight execution of queued marimo notebooks?

To automate overnight execution of marimo notebooks, you can schedule headless batch processing that generates standardized 8-cell notebooks, executes them, and records structured journal events for unattended runs.

Can I run headless batch analysis and generate morning summaries automatically?

Yes, headless batch analysis automatically processes queued designs overnight and produces a consolidated morning summary for human triage, eliminating the need for manual monitoring during execution.

How does automated self-review work for unattended notebook execution?

Automated self-review applies a deterministic protocol to executed notebooks, extracting evidence, observations, and questions, then appending these structured events to design journals for later review.

What happens if a notebook fails during scheduled batch processing?

If a notebook fails during scheduled batch processing, a 3-attempt error repair loop activates to address common issues, while package allowlist enforcement and version preflight checks prevent typical execution errors.

Does batch processing support MCP tool integration for data access?

Yes, batch processing supports MCP tool integration for data access and status updates, allowing headless marimo notebook runs to connect with external data sources during unattended overnight execution.

What is a deterministic notebook structure in headless batch execution?

A deterministic notebook structure in headless batch execution ensures standardized 8-cell marimo notebooks are generated consistently per design, enabling reliable lineage recording via tracked_pipe and verified outputs.