memory-keeper

Merge new learnings from experience records into domain knowledge summaries.

180|46|Updated Apr 12, 2026
One-click install
npx skills add https://github.com/chuanseng-ng/digital-chip-design-agents --skill memory-keeper-chuanseng-ng
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: memory-keeper
Source: https://github.com/chuanseng-ng/digital-chip-design-agents/tree/main/plugins/infrastructure/skills/memory-keeper
Command: npx skills add https://github.com/chuanseng-ng/digital-chip-design-agents --skill memory-keeper-chuanseng-ng

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill helps to distill accumulated experience records into updated domain knowledge summaries, ensuring that orchestrators have the most relevant and recent information for effective decision-making.

Core Features & Use Cases

  • Knowledge Distillation: Automatically merge new learnings from issue descriptions, applied fixes, and metric ranges into knowledge summaries.
  • Selective Domain Processing: Allows distilling knowledge for a single domain or all domains with sufficient records.
  • Flexible Min-Records Threshold: Configurable minimum number of records to trigger distillation.
  • Memory Root Management: Handles resolution and fallback of memory root locations for storing and accessing knowledge summaries.
  • Per-Project Scoping: Allows opt-out of central memory storage for per-project isolation.
  • Use Case: For a complex SoC design, after each 10 orchestrator sessions, automatically update the domain knowledge for architecture, verification, and synthesis domains, ensuring the team stays informed and can quickly iterate.

Quick Start

Distill the knowledge for the synthesis domain with a minimum of 5 records: /chip-design-infrastructure:memory-keeper --domain synthesis --min-records 5

Frequently Asked Questions about memory-keeper

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

FAQPage Schema
How do I distill domain knowledge from experience records to improve design workflow efficiency?

Distilling domain knowledge from experience records works by automatically merging new learnings from issue descriptions and applied fixes into updated knowledge summaries, ensuring orchestrators have the most recent information for effective decision-making.

What is the minimum number of experience records needed to trigger knowledge distillation?

Knowledge distillation requires a flexible minimum records threshold that you can configure per run. For example, you can set a minimum of 5 records to trigger the merging process for a specific domain.

Can I update domain knowledge summaries for a single domain instead of all domains?

Selective domain processing allows you to update knowledge summaries for a single domain or all domains with sufficient records. You can target specific areas like synthesis or verification independently.

How do I isolate domain knowledge updates for a specific project?

Per-project scoping allows you to opt-out of central memory storage to isolate domain knowledge. The system handles memory root resolution and fallback locations to keep project data separated.

What is the best way to automate experience records knowledge distillation for complex SoC design workflows?

Automating knowledge distillation for SoC design workflows involves running updates after orchestrator sessions to merge new learnings for architecture, verification, and synthesis domains, ensuring teams stay informed and iterate quickly.