Memory Management

anthropics/knowledge-work-pluginsOfficial

Finally, a memory system that treats workplace language like what it is: a decoding problem. Most assistants fumble when you say "ask Todd about the PSR for Phoenix" because they don't know Todd is Todd Martinez in Finance, PSR means Pipeline Status Report, or Phoenix is the database migration project. This builds a two-tier cache: CLAUDE.md keeps your top 30 people and common acronyms hot, while the memory/ directory holds the full glossary and detailed profiles. The tiered lookup means Claude checks the hot cache first, falls back to the full glossary, then asks you if something's still unknown. It's designed around the reality that 90% of your requests use the same 30 names and terms, but you need the long tail available when edge cases come up.

skills.sh installs: 7.6K
GitHub stars: 35.3K
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add anthropics/knowledge-work-plugins --skill memory-management --agent claude-code

Installs into .claude/skills of the current project.

Files

Memory Management

Memory makes Claude your workplace collaborator - someone who speaks your internal language.

The Goal

Transform shorthand into understanding:

User: "ask todd to do the PSR for oracle"
              ↓ Claude decodes
"Ask Todd Martinez (Finance lead) to prepare the Pipeline Status Report
 for the Oracle Systems deal ($2.3M, closing Q2)"

Without memory, that request is meaningless. With memory, Claude knows:

todd → Todd Martinez, Finance lead, prefers Slack
PSR → Pipeline Status Report (weekly sales doc)
oracle → Oracle Systems deal, not the company

Architecture

CLAUDE.md          ← Hot cache (~30 people, common terms)
memory/
  glossary.md      ← Full decoder ring (everything)
  people/          ← Complete profiles
  projects/        ← Project details
  context/         ← Company, teams, tools

CLAUDE.md (Hot Cache):