MCPServerMCP Apps server providing 6 interactive HTML applications.
## Architecture
MCP Apps are HTML/JS applications served by MCPservers that render in chat as sandboxed iframes. The server registers
claude.ai/code) when working with code in this repository.
## Project Overview
RivalSearchMCP is an MCPserver providing 9 specialized tools for web research, content discovery, and analysis. Built on FastMCP
pipeline (cache → disk → discover → replay). `cache.ts` is an in-memory session cache for the MCPserver.
- **`src/cli.ts`** — Hand-rolled arg parser, all CLI commands. No framework (no yargs/commander).
- **`src/mcp.ts`** — MCP
them.
## Project at a glance
`remindb` — token-efficient agentic memory database. Single SQLite file, MCPserver on top. Go 1.26+. Pre-1.0 beta, solo dev, no compatibility shims
there when a pattern repeats.
5. **Sosumi for Apple docs** – always use the `sosumi` MCPserver to search and fetch Apple developer documentation. Never guess at ASC API shapes, field
CLAUDE.md
Go MCPserver that gives MCP clients access to SigNoz metrics, logs, traces, alerts, dashboards,
saved views, notification channels, and docs over stdio and stateless streamable HTTP. It calls
visualize-setup`** — Generate an interactive HTML dashboard of all installed Claude Code skills and MCPservers
This repo serves two audiences: humans browsing the docs on GitHub, and Claude Code
generates Vercel AI SDK wrappers for Model Context Protocol (MCP) tools. It connects to MCPservers, discovers their tools, and generates type-safe TypeScript wrappers that can be used with
worse than carrying a duplicated sentence: a duplicated sentence costs
tokens, a missing `mcp_servers` limit or `client.beta.memory_stores.memories.update` makes the
model emit a call that fails at runtime.
### What "useless
What it is
- A mail-like layer that lets coding agents coordinate asynchronously via MCP tools and resources.
- Provides identities, inbox/outbox, searchable threads, and advisory file reservations, with human-auditable
tree over a Unix socket (wired into apps only with `-tags limoni_debug`) |
| `cmd/limoni-mcp` | MCP bridge from agents to the automation socket |
| `benchmarks` | Harness plus the cross-framework runners |
| `apps/globe
Project Overview
**mcp-trino** is a dual-purpose tool that works as both:
1. **MCPServer** - Enables AI assistants to interact with Trino's distributed SQL query engine through standardized
A file Claude Code reads at the start of every session. It holds the commands, conventions and warnings the agent needs for this project.
Where does it go?
At the repository root. Claude Code also reads CLAUDE.md files in subdirectories when it works there.
What should it contain?
Build and test commands, the project's layout, conventions that aren't obvious from the code, and mistakes to avoid. Short files tend to work better than long ones.
CLAUDE.md or AGENTS.md?
Claude Code reads CLAUDE.md; most other agents read AGENTS.md. Many projects keep one and point the other at it.