graphdo-ts: Agent for Claude Code

.github/agents/typescript-mcp-expert.agent.md

TypeScript MCP Server Expert is an agent for Claude Code from co-native-ab/graphdo-ts. It costs 32 tokens per session (1,864 once invoked), scanned A, original, MIT.

A coding assistant for building TypeScript MCP servers, programs that expose callable tools to AI applications through the Model Context Protocol. It is tailored to graphdo-ts and its Microsoft Graph, authentication, testing, and dependency-injection patterns.

In plain words
What is it for?
Use it to develop or review graphdo-ts MCP tools, Graph API access, TypeScript and Node.js code, input validation, authentication, dependency injection, and tests.
Why use it?
MCP servers combine tool definitions, validation, asynchronous code, authentication, and external API calls. This assistant helps implement those pieces according to the project's existing structure.

Agent for Claude Code

Written for Claude Code: a Claude Code subagent (agents/*.md). Also seen: model in frontmatter; mentions AGENTS.md.

This is co-native-ab/graphdo-ts's own configuration. It tells Claude Code how to work on graphdo-ts itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything graphdo-ts configures →

Needs its repository: it reads a path above its own folder, which exists only inside the repository. The line is import { GraphClient } from "../graph/client.js";.

Reuse

Borrowing it

Nothing to install: this file belongs to co-native-ab/graphdo-ts. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/co-native-ab/graphdo-ts/main/.github/agents/typescript-mcp-expert.agent.md
Clone the repo
git clone --depth 1 https://github.com/co-native-ab/graphdo-ts

Made for: Claude Code.

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 TypeScript MCP Server Expert

README.md
[![agentmods](https://agentmods.dev/badge/agents/co-native-ab/graphdo-ts/typescript-mcp-expert/github.svg)](https://agentmods.dev/agents/co-native-ab/graphdo-ts/typescript-mcp-expert)
Your own site
<a href="https://agentmods.dev/agents/co-native-ab/graphdo-ts/typescript-mcp-expert"><img src="https://agentmods.dev/badge/agents/co-native-ab/graphdo-ts/typescript-mcp-expert/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.

agentmods 80×15 button for TypeScript MCP Server Expert

Your own site · 80×15
<a href="https://agentmods.dev/agents/co-native-ab/graphdo-ts/typescript-mcp-expert"><img src="https://agentmods.dev/badge/agents/co-native-ab/graphdo-ts/typescript-mcp-expert.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 32 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,864 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
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 5.1 $0.00032 $0.01864
Opus 5.5 $0.00013 $0.00746
Sonnet 5.5 $0.00006 $0.00373
Haiku 4.5 $0.00003 $0.00186

Measured 1mo ago against content hash e3ab05f1ded3, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

TypeScript MCP Server Expert 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 1mo 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.

.github/agents/typescript-mcp-expert.agent.md · 188 lines

How it starts

The opening of the file, as written. The whole thing — 188 lines — stays where its author put it; the contents beside it link to each section on GitHub.

TypeScript MCP Server Expert (graphdo-ts)

You are a world-class expert in building Model Context Protocol (MCP) servers using the TypeScript SDK. You have deep knowledge of @modelcontextprotocol/sdk, Node.js, TypeScript, async programming, zod validation, and best practices for building robust, production-ready MCP servers.

This agent is customized for the graphdo-ts codebase. Always read AGENTS.md before making any changes — it is the authoritative source of conventions for this project.

graphdo-ts Architecture

This is a stdio-transport MCP server that wraps the Microsoft Graph API (Mail + Microsoft To Do). Key structural facts:

  • Transport: StdioServerTransport only — no HTTP, no Express, no CORS, no DNS rebinding protection needed
  • DI pattern: All dependencies are passed via ServerConfig { authenticator, graphBaseUrl, configDir, mcpServer, openBrowser } — no globals, no env vars inside tools
  • Auth: Authenticator interface (login(), token(), logout(), isAuthenticated(), accountInfo()) with three implementations: MsalAuthenticator (production), StaticAuthenticator (fixed token), MockAuthenticator (tests)
  • HTTP client: GraphClient in src/graph/client.ts — native fetch only, no Microsoft Graph SDK. All Graph calls go through client.request(method, path, body?)
  • Error types: AuthenticationRequiredError (thrown by token() when unauthenticated), GraphRequestError (from failed Graph calls — includes method, path, statusCode, code, graphMessage)
  • Tools never throw: All errors are caught and returned as { isError: true, content: [{ type: "text", text: message }] }

File Structure

src/
  index.ts         ServerConfig, createMcpServer(), main()
  auth.ts          Authenticator interface + MsalAuthenticator + StaticAuthenticator
  graph/
    client.ts      GraphClient, GraphRequestError
    types.ts       TypeScript interfaces for Graph entities
    mail.ts        getMe(), sendMail()
    todo.ts        TodoList/TodoItem CRUD + pagination
  tools/
    login.ts       login + logout tools
    mail.ts        mail_send tool
    todo.ts        todo_* tools
    config.ts      todo_config tool (human-only browser picker)
    status.ts      auth_status tool
test/
  helpers.ts       createTestEnv() — standardized mock setup
  mock-auth.ts     MockAuthenticator
  mock-graph.ts    MockState + in-memory Graph HTTP server
  integration.test.ts  Full e2e via InMemoryTransport
  graph/           Graph layer unit tests

Read the full file on GitHub · 188 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. 1mo ago First seen · 188 lines · 32 tokens per session scan A e3ab05f1ded3

Subscribe to this mod's changes

TypeScript MCP Server Expert is an agent published in the GitHub repository co-native-ab/graphdo-ts (1 stars, last pushed 2mo ago), licensed MIT. It adds 32 tokens to every session and 1,864 once invoked, about $0.0001 per session on Opus 5.5. 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.