Skip to main content
skill CX Codex

memory · harperreed — Claude Code skill: semantic memory for your agent

A skill that keeps agent notes in SQLite with embeddings and pulls them back by meaning. Useful when project context has to survive across sessions.

Install into Claude Code · Skills CLI (Vercel) All methods and platforms ↓
npx skills add harperreed/dotfiles --skill memory -a claude-code
Open on GitHub

What memory does

The skill exposes five MCP tools. mcp__memory__store writes a note with tags, mcp__memory__search runs a similarity lookup, mcp__memory__list shows recent entries, mcp__memory__delete drops one, and mcp__memory__get_context assembles context around a topic. Because matching happens on embeddings, your query wording does not have to mirror the stored text.

Everything lands in ~/.local/share/memory/memory.db, a SQLite file with embeddings kept as BLOBs; the path follows XDG_DATA_HOME. When the MCP server is not reachable, the CLI still works: memory add "..." --tag preference, memory search "...", memory list, memory export -f markdown.

Nothing gets remembered on its own. Entries appear only when a tool is called, so saving decisions and preferences is on you. Skip mcp__memory__store and the database stays empty, which leaves search with nothing to return.

What it helps with

  • carry user preferences from one session to the next
  • find an earlier decision by meaning rather than exact wording
  • load project context before starting a task
  • export accumulated notes to markdown
  • remove a stale entry from memory

Example prompts for the agent

  • Remember that I prefer TypeScript in this project and store it with the preferences tag
  • Search memory for what I said about the build setup
  • Get context on my coding preferences

How to install memory

1Skills CLI (Vercel)

npx skills add harperreed/dotfiles --skill memory -a claude-code

2Manually: put the files in place

~/.claude/skills/memory/SKILL.md

Copy the whole skill folder; for one project use .claude/skills/memory/

Step by step, with commands for each platform: How to install skills in Claude Code: five ways

Questions about memory

How do I install memory in Claude Code?

Run npx skills add harperreed/dotfiles --skill memory -a claude-code. The skill lands at ~/.claude/skills/memory/SKILL.md.

What does the memory skill do?

It gives the agent vector memory: tagged note storage, semantic search over those notes, listing recent entries and deletion. All of it sits in a local SQLite database.

Where is the memory data stored?

In ~/.local/share/memory/memory.db. The location respects XDG_DATA_HOME, and embeddings live in the same file as BLOBs.

Similar skills

All skills for Agent setup: memory, context, orchestration →

memory by other authors comes first.

memory OrchestKit Search a knowledge graph for earlier decisions, load session context, and inspect decision history. Use it when you need to recover what a project has already settled. CX Codex +1 749installs memory johnlindquist Connects your coding agent to the basic-memory CLI to store notes, run semantic searches, and keep project context across sessions. CX Codex +1 362installs memory plasma-ai A skill that sets the discipline for a node's persistent memory: what to write, how to split pages, when to prune. Useful when your agent hoards history instead of knowledge. CX Codex +1 ★ 786repo stars setup-matt-pocock-skills Matt Pocock This skill configures a repository for Matt Pocock's engineering skills by setting its issue tracker, triage labels, and domain docs layout. Run it before using the other skills. 1,052,851installs ask-matt Matt Pocock Routes you to the right Matt Pocock skill or flow for the task at hand. Use it when you need a starting point for a feature, prototype, incoming issue, or hard bug. 766,587installs to-tickets Matt Pocock Turns a plan, spec, or conversation into small end-to-end tickets with explicit blockers. Use it to agree on the work breakdown before tickets go to a tracker. 711,230installs writing-plans Jesse Vincent This skill turns requirements for a multi-step task into an implementation plan before code changes begin. It helps when another engineer needs exact files, interfaces, and checks. CX Codex 282,859installs caveman-review JuliusBrussee Turns code review comments into one line per finding: location, problem, fix. Useful when a diff or PR needs a fast pass without the filler. CX Codex +1 411,800installs caveman-compress JuliusBrussee Compresses CLAUDE.md, todo lists, and other prose files into a shorter format while retaining a readable backup. CX Codex +1 411,500installs

From the same set8

Project sponsors

Companies that keep this analytics open
Gold sponsor

Impressions, clicks and code copies — open statistics. We keep the numbers public for the same reason NeBlask itself is public. Become a sponsor