lattice-goal-md

Author GOAL.md and MEMORY.md sections for LATTICE fitness functions and ratchet workflows.

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

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps you maintain reliable, measurable GOAL.md and MEMORY.md files so the LATTICE meta-harness can score progress, run improvement loops, and safely hand off sessions without drifting.

Core Features & Use Cases

  • Fitness-function authoring: Define measurable health criteria mapped to runnable commands and section scoring scripts.
  • Improvement-loop specification: Encode the scoring-before/propose/apply/score-after/accept-reject/log workflow with concurrency control.
  • Action catalog and operating mode: Provide immediately executable commands plus startup, failure modes, and fallback behavior.
  • MEMORY tracking: Record failed experiments, resolve open decisions, and document session handoff notes for continuity.

Quick Start

Ask the agent to update or create the appropriate section GOAL.md and its related MEMORY.md entries to match the latest scripts/score-<section>.sh behavior.

Frequently Asked Questions about lattice-goal-md

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

FAQPage Schema
How do I define measurable fitness functions for a ratchet workflow?

Measurable fitness functions are defined in GOAL.md by aligning health criteria to runnable commands and section scoring scripts, ensuring the ratchet workflow produces deterministic, reproducible outputs.

What is the best way to maintain context across LATTICE meta-harness work sessions?

Maintaining context across work sessions requires updating MEMORY.md with failed experiments, resolved decisions, and handoff notes, ensuring the LATTICE meta-harness safely resumes without drifting.

How do I encode an improvement loop with concurrency control in GOAL.md?

Encoding an improvement loop involves specifying the scoring-before, propose, apply, score-after, accept-reject, and log workflow steps deterministically in GOAL.md, including concurrency control rules.

Can I add new sections to a ratchet workflow without breaking harness reproducibility?

Adding new sections requires authoring a section GOAL.md mapped to runnable scripts/score-<section>.sh outputs, ensuring the ratchet workflow maintains harness reproducibility and consistent scoring behavior.

When should I refine existing fitness function criteria in a LATTICE meta-harness?

Refining existing fitness function criteria is necessary when scoring reveals gaps in the LATTICE meta-harness, requiring updates to GOAL.md to realign criteria with runnable commands and deterministic outputs.

Why does my LATTICE meta-harness drift during session handoffs?

Harness drift during session handoffs occurs when MEMORY.md is not updated with rejection, decision, and handoff rules, causing the LATTICE meta-harness to lose context and fail reproducibility.