Curated, modular memory and instruction files for Claude Code, maintained by Ribose for developers across our orgs (riboseinc/, metanorma/, lutaml/, plurimath/, relaton/*).

Audience signposts

Quick start

Install one file via curl (zero setup required):

curl -fsSL https://raw.githubusercontent.com/riboseinc/claude-memory-files/main/tools/install.sh \
  | bash -s -- instructions/github-pr-title-issue-link.md

For the durable workflow (manage installs, list outdated, submit your own), install the companion plugin once:

/plugin marketplace add riboseinc/claude-memory-files
/plugin install memory-files-tools@claude-memory-files

Then /list-memory-files to browse, /install-memory-file <slug> to install. See Using the slash commands.

Five-level learning path

  1. Install one file — curl one-liner, verify it loads in your next Claude Code session.

  2. Install several — mix universal and team scope files; see scope rubric.

  3. Customise a personal-share — walkthrough using user-maintained-gems.

  4. Submit your own — guide for /submit-memory-file.

  5. Maintain a category — tag-team review with the maintainer; help curate seed batches.

What this repo deliberately does NOT host

Slash commands, subagents, skills, MCP server configs, output styles, and keybindings are abundantly catalogued by awesome-claude-code and similar aggregators. We focus on the white-space categories: memory, instructions, settings fragments, project CLAUDE.md templates, and path-scoped rules. See the README "Further reading" for the curated aggregator list.

Site sections

  • Getting started — installation walkthrough.

  • Concepts — design rationale: scope rubric, mechanism/data, instructions vs memory, settings-fragments, hooks-deferred.

  • Guides — task-based walkthroughs.

  • Reference — frontmatter schema, file types, validator rules, manifest format, slash commands.

  • Roadmap — v1 design, update regulation, v1.1 and v2 candidates.

Design history