LATTICE Global Ratchet Orchestration

Select the lowest-scoring LATTICE section, run autoresearch, and apply accepted proposals.

3|Updated May 11, 2026
One-click install
npx skills add https://github.com/JeromyJSmith/lattice-platform --skill lattice-global-ratchet-orchestration
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: LATTICE Global Ratchet Orchestration
Source: https://github.com/JeromyJSmith/lattice-platform/tree/main/.agents/skills/lattice-global
Command: npx skills add https://github.com/JeromyJSmith/lattice-platform --skill lattice-global-ratchet-orchestration

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It coordinates an auditable, reproducible improvement cycle that targets the lowest-performing section of the LATTICE platform while ensuring progress is measurable and globally beneficial.

Core Features & Use Cases

  • Global improvement loop: averages all seven section scores, identifies the weakest section each cycle, and drives the ratchet toward higher global mean.
  • Evidence-backed auditing: records pre/post global health snapshots, cross-section decisions, and proposal outcomes into lattice/harness/* Pixeltable tables for replayability.
  • Safety guardrails: halts and logs a ratchet.error event if a section drops by more than 10 points, preventing untracked or harmful regressions.
  • Deterministic orchestration workflow: runs the appropriate section autoresearch via the harness bootstrap runner and applies only accepted proposals using git apply in a sandbox-safe manner.

Quick Start

Run the global ratchet by invoking the autoresearch cycle for the lowest-scoring section using bash meta/harness/bootstrap/run-autoresearch.sh <lowest-section>.

Frequently Asked Questions about LATTICE Global Ratchet Orchestration

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

FAQPage Schema
How do I orchestrate a reproducible improvement loop for the weakest section of a platform?

Global ratchet orchestration automates a reproducible improvement loop by selecting the lowest-scoring section, running an autoresearch cycle, and applying only accepted proposals to ensure auditable progress.

How do I record audit trails for automated code improvement proposals?

Record audit trails for automated code improvement by persisting pre/post health snapshots, cross-section decisions, and proposal outcomes into Pixeltable-backed tables to ensure global progress remains replayable.

What's the best way to prevent regression drift when applying automated research proposals?

Prevent regression drift during automated research by enforcing safety guardrails that halt the ratchet and log a ratchet.error event if any section score drops by more than 10 points.

How do I apply accepted autoresearch proposals safely in a sandbox environment?

Apply accepted autoresearch proposals safely by invoking the harness bootstrap runner and using git apply within a sandbox-safe environment to ensure deterministic orchestration workflow.

Does the global ratchet orchestration workflow require immutable baseline anchoring?

Yes, global ratchet orchestration requires immutable baseline anchoring and single-writer lock enforcement to coordinate auditable, reproducible improvement cycles targeting the lowest-performing sections.