Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add instructions/dongnh/matter_webcontrol/claude-mdgit clone --depth 1 https://github.com/dongnh/matter_webcontrolWrote 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/dongnh/matter_webcontrol/claude-md)<a href="https://agentmods.dev/instructions/dongnh/matter_webcontrol/claude-md"><img src="https://agentmods.dev/badge/instructions/dongnh/matter_webcontrol/claude-md.svg" alt="Measured on agentmods" 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 | $0.01381 | $0.01381 |
| Opus 5 | $0.00691 | $0.00691 |
| Sonnet 5 | $0.00276 | $0.00276 |
| Haiku 4.5 | $0.00138 | $0.00138 |
Grade B, and why
matter_webcontrol CLAUDE.md scanned grade B with 1 finding 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 4d 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.
Asks for rootmediumPrivilege escalation
A mod that escalates privileges can change anything on the machine, not only the project.
# HTTP server — no sudo needed when devices are already on the LAN. How it starts
The opening of the file, as written. The whole thing — 86 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Matter Web Controller — Agent Guide
What is this?
A Python server for controlling Matter smart home devices (lights, sensors) and external logical bridges via REST API and MCP.
Project structure
cli/
core.py — DeviceController: all business logic (single source of truth)
constants.py — magic values: Thermostat modes/clusters, sensor keys, mired bounds, extract_matter_pin
conversions.py — pure unit math (brightness/mired/kelvin/centi); zero deps
serializers.py — pure (device, names) -> dict builders + resolved_names merge
paths.py — resolve one data dir + every cache/storage path
schemas.py — Pydantic request models (POST bodies)
auth.py — constant-time api-key check + non-loopback bind refusal
server.py — FastAPI routes only (thin wrappers calling core)
mcp_server.py — MCP server (HTTP client calling matter-srv)
matter_bridge.py — Matter protocol bridge (WebSocket) + public facade + persistence
logic_bridge.py — Logical bridge manager (HTTP federation with remote instances)
tests/ — pytest suite; fakes.py is the single fake source (shared with dev/)
Architecture rules
- All business logic lives in
cli/core.py—server.pyandmcp_server.pyare thin wrappers only. Never duplicate logic. - MCP server is an HTTP client — It calls the running
matter-srvHTTP server. It does NOT importcore.py,matter_bridge.py, orlogic_bridge.py. - Device IDs are canonical — Always
dev_*format (e.g.dev_a3f7c1b2). Aliases are display-only, never used for ID resolution. - Physical vs logical routing — Commands check logical bridges first, then fall back to physical Matter devices. There is exactly one router (
DeviceController._route) and one deduped enumerator (._iter_devices) — route/enumerate through them, don't re-implement the traversal. - Bridge access goes through the facade —
core/servernever touchbridge._privatemembers; use the public methods onMatterBridgeServer(sync,names_for,add_alias,device_ids_for_node,subscribe_occupancy, …). - Mutations are POST-only with a Pydantic body from
cli/schemas.py; secrets go in the body, never the URL.
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.
- 4d ago First seen · 86 lines · 1,381 tokens per session scan B a7464f4650cc
matter_webcontrol CLAUDE.md is an instructions file published in the GitHub repository dongnh/matter_webcontrol (0 stars, last pushed 24d ago), licensed MIT. It adds 1,381 tokens to every session, about $0.0069 per session on Opus 5. A static security scan graded it B with 1 finding (asks for root). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other instructions, from other repositories
cml-mcp AGENTS.md
AGENTS.md instructions for xorrkaz/cml-mcp, covering agents.md — cml mcp server, project overview, compatibility goal, repository layout and tool modules (src/cmlmcp/tools/).
brilliant_sdk AGENTS.md
Instructions for brilliantlabsAR/brilliant_sdk, covering brilliant sdk — agent guide, how an app works (the pattern behind everything), minimal reading paths, verify without hardware and testing.
NeoMind CLAUDE.md
Claude Code instructions for camthink-ai/NeoMind, covering neomind — edge ai platform for iot, development commands, ecosystem repositories, extension package contract (.nep) and device type template contract (json).
cad-cae-copilot copilot-instructions.md
Copilot instructions for armpro24-blip/cad-cae-copilot, covering github copilot — aieng workspace and essentials.
phone-mcp CLAUDE.md
Claude Code instructions for premex-ab/phone-mcp, covering claude.md, project overview, build commands, architecture and module layout.
zmk-config AGENTS.md
AGENTS.md instructions for urob/zmk-config, covering customization guide, ground rules, how the multi-board layout works, adding a new board and where to change what.