Borrowing it
Nothing to install: this file belongs to cacack/mcp-server-zwave-js-ui. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/cacack/mcp-server-zwave-js-ui/main/CLAUDE.mdgit clone --depth 1 https://github.com/cacack/mcp-server-zwave-js-uiWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/instructions/cacack/mcp-server-zwave-js-ui/claude-md)<a href="https://agentmods.dev/instructions/cacack/mcp-server-zwave-js-ui/claude-md"><img src="https://agentmods.dev/badge/instructions/cacack/mcp-server-zwave-js-ui/claude-md/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/instructions/cacack/mcp-server-zwave-js-ui/claude-md"><img src="https://agentmods.dev/badge/instructions/cacack/mcp-server-zwave-js-ui/claude-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.01483 | $0.01483 |
| Opus 5 | $0.00741 | $0.00741 |
| Sonnet 5 | $0.00297 | $0.00297 |
| Haiku 4.5 | $0.00148 | $0.00148 |
Grade A, and why
mcp-server-zwave-js-ui CLAUDE.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 9d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 58 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
Project Overview
Python MCP server for Z-Wave device management. It connects to the
zwave-js-server WebSocket exposed by Z-Wave JS UI
(the "Z-Wave JS Server" port, default 3000 — not the 8091 web UI) and
exposes the Z-Wave network as MCP tools.
Scope is delivered in levels:
- Read-only (done) — controller info, node listing/detail, values, config parameters.
- Read/write (done) — set values, set config parameters, set node name/location, manage associations.
- Admin/lifecycle (mostly done) — inclusion/exclusion, re-interview, network heal, remove failed node, plus rebuild/firmware status polls. Triggering OTA firmware still pending (see below).
Read-only lockdown: setting ZWAVE_JS_READ_ONLY (1/true/yes/on) makes server._apply_read_only() remove every tool in server._MUTATING_TOOLS from the FastMCP registry at import, so a locked server never advertises them (the pattern used by mcp-server-vyos). Any new mutating tool must be added to _MUTATING_TOOLS — a test asserts the mutating and read-only name sets partition the registry.
Commands
uv sync --extra dev # create .venv from uv.lock (matches CI's --locked)
uv run pytest
uv run ruff check . && uv run ruff format --check .
ZWAVE_JS_URL=ws://<host>:3000 uv run python -m zwave_js_ui_mcp # run server (stdio)
Architecture
src/zwave_js_ui_mcp/server.py— MCP tool registration (FastMCP). Tools are thin: connect, delegate to a projection, return the dict.src/zwave_js_ui_mcp/client.py— the entire backend.connected_driver()opens/tears down a connection;project_*functions turn library model objects into plain JSON-serializable dicts.
The tool layer never touches a zwave-js-server-python object — only the dicts client.py returns. This isolates the transport so it can be swapped without touching tools.
Key Design Decisions
- Depends on
zwave-js-server-python(the Home Assistant library) for connection, schema negotiation, and — decisive for level 3 — firmware/heal progress events. This pins Python >= 3.12 (the library's floor). - Connection is per tool call (open → wait for full-state dump → read → close), matching the stateless MCP style.
client.listen()must run as a background task fordriver_readyto fire and forasync_send_commandfutures to resolve. - Mutating ops fit the per-call model because they are single request→response or fire-and-forget:
set_value/set_configvalidate against live metadata then wait for the command result; inclusion/exclusion and network heal only toggle server-side controller state (a laterstop_*call ends them). Long-running ops surface progress by a poll tool (zwave_rebuild_routes_status,zwave_firmware_update_status) rather than event streaming. Triggering an OTA firmware update does not fit — flashing streams progress over minutes, longer than one tool call, so it's deferred until a persistent-connection design lands. Interactive S2 inclusion (DSK/PIN grant over events) likewise can't complete statelessly;begin_inclusiondocuments this. - Configuration-CC (112) values are excluded from
zwave_node_valuesand surfaced only byzwave_node_config, so parameters aren't reported twice. - Server URL via
ZWAVE_JS_URL(defaultws://localhost:3000). The WS server needs no auth by default; only the 8091 UI does.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 9d ago First seen · 58 lines · 1,483 tokens per session scan A eb9d87fa6f59
mcp-server-zwave-js-ui CLAUDE.md is an instructions file published in the GitHub repository cacack/mcp-server-zwave-js-ui (0 stars, last pushed 1mo ago), licensed MIT. It adds 1,483 tokens to every session, about $0.0074 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other instructions, from other repositories
next.js AGENTS.md
AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.
codex AGENTS.md
AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.
vscode buildNext.instructions.md
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).
langchain AGENTS.md
AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.
deepseek-harness AGENTS.md
AGENTS.md instructions for deepseek-ai/deepseek-harness, covering agents.md, pre-stable apis and released session data, repository layout, commands and host sandbox failures.