negotiation, result envelopes, error codes. It defines functions only; sourcing
it starts nothing. `run_mcp_server` is the stdio entry point.
- `mcpserver_http.sh` — the Streamable HTTP binding. Parses HTTP, enforces
Common mistake: omitting the artist folder.
First-time setup: `cp config/config.example.yaml ~/.bitwize-music/config.yaml` — see `config/README.md`.
---
## MCPServer — Preferred Data Access
The `bitwize-music-mcp` server is the **preferred way to query
MCP is a VSCode extension that enables searching, managing, and installing Model Context Protocol (MCP) servers. It extends GitHub Copilot Chat's capabilities by providing a chat participant interface
MCP gateway. It allows users to create and manage multiple playbooks. Each playbook aggregates MCPservers and prompts.
We've just migrated director from a single tenant, local first architecture
fact.
## Shared code intelligence
Both agents query the same index through the `flyto-indexer` MCPserver —
registered in `.mcp.json` for Claude and `~/.codex/config.toml` for Codex.
- A `post-commit` hook reindexes
answers.
GrayMatter is a single-binary Go memory system for AI agents. Library + CLI +
MCPserver + TUI. Facts persist to bbolt with optional vector embeddings, for
roughly
HTTP calls with direct Obsidian plugin API calls. This allows reuse of all existing MCPserver logic while gaining performance benefits.
## Code Quality Guidelines
### SOLID Principles Application
- **Single Responsibility**:
- `ObsidianAPI
Project Overview
This is the AppleScript MCP (Model Context Protocol) Server - a dual-implementation MCPserver that allows AI assistants to execute AppleScript on macOS systems. The server is available
with Tailwind
bunx @modelcontextprotocol/inspector # Test MCP tools locally
bun run headless --root # Headless .bbmodel MCPserver over stdio (no Blockbench)
bun run build:headless # Bundle it to dist/headless/blockbench-mcp-headless.js
been written yet this session.
## MCP Steroid — IntelliJ IDE Control
When the `localhost-6315` MCPserver (mcp-steroid) is connected, its skill guides are exposed as MCP resources and fetched
goal?"** — For any activity
2. **"Did this increase leverage?"** — For any output
---
## Part 11: MCPServers
### Connected Servers
| Server | Status | What It Enables |
|--------|--------|-----------------|
| Gmail | Connected | Email triage, drafting |
| Google Calendar
before the server so `server/internal/api/ui/dist` contains the files required by `go:embed`.
## Core invariants
- Node IDs are deterministic SHA-256 identities. Config and MCP collection must emit the same MCPServer
this repo until that is done.
You have access to a `memstack-skills` MCPserver. This is your **PRIMARY** source for task-specific guidance. ALWAYS check memstack-skills via `find
boost rules ===
## Laravel Boost
- Laravel Boost is an MCPserver that comes with powerful tools designed specifically for this application. Use them.
## Artisan
- Use the `list-artisan-commands` tool when
healthcare plugin — conventions
- **Local data**: skills and bundled MCPservers write to `~/.claude/data/healthcare/ /`, where the component is the *server's* name, not the skill's (the contracts skill's server
github.copilot.chat.agentDebugLog.fileLogging.enabled`
### MCP Across Agent Types
- **MCP support in Copilot CLI and Claude agents** - MCPservers registered in VS Code are now bridged into Copilot CLI and Claude agent workflows, including
working in this repository.
This is the **upstream source** for `odoo-mcp` — the Odoo MCPserver published
to PyPI as `odoo-mcp` (`uvx odoo-mcp`) and consumed by downstream projects
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.