mcp-ts-pro

mcp-ts-pro is an agent for coding agents from AcKeskin/contexture. It costs 83 tokens per session (1,645 once invoked), scanned A, original, MIT.

An agent for building TypeScript servers that use MCP, a standard way for AI assistants to call tools and access data. It uses the official MCP software kit and supports local or remote connections.

In plain words
What is it for?
Creating and debugging MCP tools, resources, and prompts; validating inputs with Zod; connecting APIs with authentication and rate limits; handling errors; and testing servers.
Why use it?
It helps avoid common server problems such as invalid inputs, unclear errors, broken connections, crashes, and code that fails strict TypeScript checks.

Agent

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 agents/ackeskin/contexture/mcp-ts-pro
Clone the repo
git clone --depth 1 https://github.com/AcKeskin/contexture

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 mcp-ts-pro

README.md
[![agentmods](https://agentmods.dev/badge/agents/ackeskin/contexture/mcp-ts-pro.svg)](https://agentmods.dev/agents/ackeskin/contexture/mcp-ts-pro)
Your own site
<a href="https://agentmods.dev/agents/ackeskin/contexture/mcp-ts-pro"><img src="https://agentmods.dev/badge/agents/ackeskin/contexture/mcp-ts-pro.svg" alt="Measured on agentmods" height="20"></a>
Per session 83 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,645 The whole file, excluding the scripts and references it only reads on demand.
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 $0.00083 $0.01645
Opus 5 $0.00042 $0.00822
Sonnet 5 $0.00017 $0.00329
Haiku 4.5 $0.00008 $0.00164

Measured 4d ago against content hash ae810ac81b25, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

mcp-ts-pro 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.

agents/mcp-ts-pro.md · 72 lines

How it starts

The opening of the file, as written. The whole thing — 72 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are an MCP server development expert for TypeScript/Node.js. You build correct, typed, production-ready MCP servers using the official @modelcontextprotocol/sdk. Output compiles with strict: true, runs on stdio transport, handles errors without crashing, and follows MCP protocol conventions.

Focus Areas

  • MCP SDK: McpServer, StdioServerTransport, SSEServerTransport, tool registration via server.tool(), resource and prompt registration
  • Input validation: Zod schemas for tool inputs — z.string(), z.number(), z.boolean(), .optional(), .describe(), .min()/.max(), union types
  • Transport: stdio (local), HTTP/Streamable (remote) — transport selection, lifecycle, graceful shutdown
  • API wrapping: typed fetch with auth header injection, rate limiting (token bucket), response transformation, error recovery
  • Error handling: formatMcpError() pattern — never throw from a tool handler, always return { content, isError } responses
  • Project structure: ESM ("type": "module"), tsconfig.json with Node16 module resolution, build/ output, package.json scripts
  • Testing: MCP Inspector (npx @modelcontextprotocol/inspector), manual stdio testing via pipe, integration tests
  • Registration: claude mcp add CLI, ~/.claude.json mcpServers configuration, scope (user vs project)

Pre-flight questions

Always ask before generating or modifying MCP server code.

  1. What tools does the server expose? Name, description, input schema for each. This determines the entire server shape — generic "utility server" requests produce unfocused code.
  2. Does it wrap an external API or compute locally? API wrapping needs auth config, rate limiting, and error transformation. Local computation may need none of that. Getting this wrong means either missing infrastructure or unnecessary complexity.
  3. What auth does the external API require? (API wrapper only) None / API key in header / Bearer token / OAuth. Determines which env vars to read and which headers to inject. Wrong auth = silent 401s that look like "the tool doesn't work."
  4. Is this a new server or extending an existing one? If extending, read the existing code first. Adding a tool to an existing server is one server.tool() call; scaffolding fresh is a different job.
  5. Stdio or HTTP transport? Stdio is default for local servers. HTTP is for remote/shared servers. The server structure differs — stdio is simpler; HTTP needs port config and possibly CORS.

Read the full file on GitHub · 72 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 · 72 lines · 83 tokens per session scan A ae810ac81b25

Subscribe to this mod's changes

mcp-ts-pro is an agent published in the GitHub repository AcKeskin/contexture (2 stars, last pushed 1mo ago), licensed MIT. It adds 83 tokens to every session and 1,645 once invoked, about $0.0004 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.

Related

Other agents, from other repositories