pieces-ltm

Retrieve nine months of Pieces LTM context for feature or bug work.

1|Updated Nov 25, 2025
One-click install
npx skills add https://github.com/89jobrien/dotfiles --skill pieces-ltm
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: pieces-ltm
Source: https://github.com/89jobrien/dotfiles/tree/main/dot-claude/skills/pieces-ltm
Command: npx skills add https://github.com/89jobrien/dotfiles --skill pieces-ltm

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Use before starting work on a feature or bug to get historical context from 9 months of LTM. Use when you need to know what you tried before, why a decision was made, or what was the state of X last month.

Core Features & Use Cases

  • Retrieve long-term memory context (9 months) to inform current work.
  • Enable decision archaeology by surfacing prior rationale and outcomes.
  • Provide context for planning, risk assessment, and feature scoping.

Quick Start

Ask Claude to load the current project’s Pieces LTM context before you start a feature or bug.

Frequently Asked Questions about pieces-ltm

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

FAQPage Schema
How do I retrieve historical context from past development work before starting a new feature?

To retrieve historical context, you can load nine months of Pieces LTM data. This surfaces prior decisions and past project states to inform your current feature or bug work.

What is decision archaeology and how does long-term memory help investigate prior code decisions?

Decision archaeology uses long-term memory to surface the rationale behind prior development choices. By querying nine months of LTM context, you can understand why specific decisions were made.

How do I get context about the state of a project from last month for bug investigation?

You can retrieve the previous state of a project by querying historical LTM context. This provides up to nine months of historical data to guide your current bug investigation.

Do I need access to the Pieces health endpoint to use Claude for historical context retrieval?

Yes, retrieving historical context requires access to the Pieces health endpoint. Claude uses this endpoint and its querying capabilities to fetch LTM data with deterministic behavior and clear failure handling.

When should I load long-term memory context for software engineering tasks?

You should load long-term memory context when starting new work, investigating prior decisions, or needing context about past project states. It provides historical data for planning and risk assessment.

What are the limitations of using historical context retrieval for feature scoping?

The historical context retrieval is limited to a nine-month window of LTM data. While it provides deterministic behavior and clear failure handling, it cannot surface decisions or project states older than nine months.