By ryanthedev
Persistent memory for LLMs with search, multi-source docs, and dreaming. SQLite FTS5, markdown storage, git-tracked history.
Install or update grug-brain. Downloads prebuilt binary or builds from source.
Review and maintain grug-brain memories. Commits changes to git, finds cross-links, flags stale memories. Use with /loop for periodic maintenance.
Add documentation to grug-brain for FTS-indexed search. Ingest from local directories or GitHub repos. Supports subfolder paths like github:owner/repo/path/to/section.
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Persistent memory for LLMs. Point it at any number of directories, and grug indexes every markdown file into one FTS5 search index. Git sync keeps brains in lockstep across machines. Conflicts get caught, not lost.
claude plugin add grug-brain
/install
/install builds from source, installs the background service, configures your brains, and verifies everything works. Run it again after plugin updates to pick up the new version.
Once the server is running, open http://localhost:7777 (or check ~/.grug-brain/serve.port if the default port was taken).
Brain switcher, category tree, memory list with live search, DOMPurify-sanitized markdown preview, and a global similarity graph (sigma.js + Fruchterman-Reingold layout). Live-reloads via SSE when any file changes on disk. Theme follows OS preference and can be toggled manually.
Every memory opens in edit mode by default:
[[wikilink]] and #tag syntax highlighting.[[ to complete from memory titles; type # to complete from the tag index.PUT /api/memory/{path} with ETag concurrency control.beforeunload for browser-tab close.+ next to any category, or use Cmd-K "New memory". First save prompts for a filename.rewrite_links=true by default rewrites every [[wikilink]] across the brain atomically.Fuzzy quick-switcher over memories, categories, and commands:
| Command | Action |
|---|---|
| Toggle theme | Cycle light / dark / system |
| New memory | Open a blank draft editor |
| Rename | Open rename modal for current memory |
| Delete | Open delete confirm modal for current memory |
| Jump to category | Navigate to the first category |
| Shortcut | Surface | Action |
|---|---|---|
| Cmd-S / Ctrl-S | Anywhere | Save current memory |
| Cmd-K / Ctrl-K | Anywhere | Open / close Cmd-K palette |
| Escape | Any open modal or palette | Close / dismiss |
| Enter / Space | Memory list item | Open memory |
| Arrow Up / Down | Palette list | Navigate results |
| Enter | Palette | Dispatch selected item |
| Tab / Shift-Tab | Any modal | Cycle focus within modal (trapped) |
All modals restore focus to the element that was active when the modal opened.
Manual screen-reader verification — expected announcements with VoiceOver on macOS (Safari):
Edit save flow:
Conflict modal:
Palette navigate:
Delete confirm modal:
Total web/vendor/ size: 1,035,743 bytes (1.01 MB)
19 software engineering skills from Code Complete, APOSD, GoF, and Clean Architecture. Skills are internal (slash-invocable; injected via Read() — not auto-triggered). Research → plan → build workflow with Gate-field adaptive gates (Full | Standard | Minimal) and per-phase orchestrated commits.
Discipline plugins for Claude Code — prompt engineering, agent dispatch, writing, search, and browser/DevTools control
System design interview preparation and architecture review — structured frameworks for distributed systems
React Native foundation skills — documentation search, diagnosis, debugging, design-first coding, and development tooling
npx claudepluginhub ryanthedev/grug-brain.mcpGive your AI a memory — mine projects and conversations into a searchable palace. 33 MCP tools, auto-save hooks, and guided setup.
MCP server that saves 98% of your context window with session continuity. Sandboxed code execution in 11 languages, FTS5 knowledge base with BM25 ranking, and automatic state restore across compactions.
Upstash Context7 MCP server for up-to-date documentation lookup. Pull version-specific documentation and code examples directly from source repositories into your LLM context.
Intelligent draw.io diagramming plugin with AI-powered diagram generation, multi-platform embedding (GitHub, Confluence, Azure DevOps, Notion, Teams, Harness), conditional formatting, live data binding, and MCP server integration for programmatic diagram creation and management.
Build and maintain an LLM-curated personal knowledge base in your project — Andrej Karpathy's LLM Wiki pattern, designed to scale to thousands of pages without becoming a context bottleneck. Now with an optional compiled graph layer for typed, provenance-backed relationships.
Comprehensive skill pack with 66 specialized skills for full-stack developers: 12 language experts (Python, TypeScript, Go, Rust, C++, Swift, Kotlin, C#, PHP, Java, SQL, JavaScript), 10 backend frameworks, 6 frontend/mobile, plus infrastructure, DevOps, security, and testing. Features progressive disclosure architecture for 50% faster loading.