scrypted-mcp-plugin CLAUDE.md

scrypted-mcp-plugin CLAUDE.md is an instructions file for coding agents from sv-tools/scrypted-mcp-plugin. It costs 3,499 tokens per session, scanned A, original, MIT.

Project instructions for a Scrypted plugin that provides an MCP endpoint, allowing AI clients such as Claude Desktop or Claude Code to access logs, plugins, and devices on a Scrypted server.

In plain words
What is it for?
Use it when building, formatting, linting, testing, deploying, or publishing the Scrypted MCP plugin.
Why use it?
It gives developers the commands, architecture, authentication details, and continuous-integration checks needed to work on the plugin.

Instructions file

Install

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.

agentmods
npx agentmods add instructions/sv-tools/scrypted-mcp-plugin/claude-md
Clone the repo
git clone --depth 1 https://github.com/sv-tools/scrypted-mcp-plugin

Wrote 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.

agentmods badge for scrypted-mcp-plugin CLAUDE.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/sv-tools/scrypted-mcp-plugin/claude-md.svg)](https://agentmods.dev/instructions/sv-tools/scrypted-mcp-plugin/claude-md)
Your own site
<a href="https://agentmods.dev/instructions/sv-tools/scrypted-mcp-plugin/claude-md"><img src="https://agentmods.dev/badge/instructions/sv-tools/scrypted-mcp-plugin/claude-md.svg" alt="Measured on agentmods" height="20"></a>
Per session 3,499 This file is loaded in full into every session.
When invoked 3,499 The same file — it is already loaded in full.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.03499 $0.03499
Opus 5 $0.01750 $0.01750
Sonnet 5 $0.00700 $0.00700
Haiku 4.5 $0.00350 $0.00350

Measured 4d ago against content hash a434aec24977, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

scrypted-mcp-plugin 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 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.

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.

CLAUDE.md · 108 lines

How it starts

The opening of the file, as written. The whole thing — 108 lines — stays where its author put it; the contents beside it link to each section on GitHub.

scrypted-mcp

Scrypted plugin that exposes a Streamable HTTP Model Context Protocol endpoint. AI clients (Claude Desktop, Claude Code, etc.) authenticate via OAuth backed by Scrypted user accounts and call MCP tools to inspect logs, manage plugins, and query devices on the same Scrypted server.

Commands

npm run build       # scrypted-webpack → out/main.nodejs.js (+ plugin.zip)
npm run fmt         # prettier --write
npm run fmt:check   # CI gate
npm run lint        # eslint
npm run lint:fix
npm run scrypted-deploy-debug   # build + push to a running Scrypted server (see SDK docs)

CI runs fmt:check, lint, build on every push/PR (.github/workflows/ci.yml). publish.yml runs on v* tags and pushes to npm.

Architecture

The plugin is one process inside Scrypted. src/main.ts declares ScryptedMcpPlugin extends ScryptedDeviceBase implements HttpRequestHandler and registers the full MCP tool surface on a per-session McpServer. Inbound HTTP requests are dispatched in onRequest:

  1. OAuthService.handle() claims /authorize, /token, /register, and the four .well-known/* documents (oauth-protected-resource, oauth-authorization-server, openid-configuration, jwks.json). If it returns true, the response is already populated.
  2. Otherwise, if the path resolves to /mcp, we verify the Bearer JWT, look up (or create) a WebStandardStreamableHTTPServerTransport keyed by the Mcp-Session-Id header, bridge the Scrypted request to a Web-standard Request, and forward.
  3. Anything else returns 404.

src/http-bridge.ts is the only adapter between Scrypted's HttpRequest/HttpResponse and Web-standard Request/Response. Streamed responses (Content-Type: text/event-stream) flow through HttpResponse.sendStream(AsyncGenerator<Buffer>); everything else buffers and sends.

Two SSE behaviours are inherited from the MCP SDK transport (since 1.30.0) rather than implemented here, and both depend on the bridge staying byte- and header-transparent:

Read the full file on GitHub · 108 lines

Changes

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.

  1. 4d ago First seen · 108 lines · 3,499 tokens per session scan A a434aec24977

Subscribe to this mod's changes

scrypted-mcp-plugin CLAUDE.md is an instructions file published in the GitHub repository sv-tools/scrypted-mcp-plugin (5 stars, last pushed 6d ago), licensed MIT. It adds 3,499 tokens to every session, about $0.0175 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.