What is mcp-protocol-expert?

A guide to building and connecting Model Context Protocol (MCP) servers, which let AI applications call tools and access structured information. It covers message handling, tool input schemas, errors, versions, and protocol compliance.

Use it to build MCP servers, define tool parameters, implement standard communication, validate requests and responses, handle errors, and connect MCP services to applications such as Claude.

It helps make AI tools understandable and reliable for language models and compatible with the MCP specification. It also clarifies when something should be exposed as a tool or as a resource.

How jsavin/iac-mcp configures Claude Code

Reuse

Borrowing it

Nothing to install: this file belongs to jsavin/iac-mcp. 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/jsavin/iac-mcp/main/.claude/agents/mcp-protocol-expert.md

Clone the repo

git clone --depth 1 https://github.com/jsavin/iac-mcp

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.

README.md

[![agentmods](https://agentmods.dev/badge/agents/jsavin/iac-mcp/mcp-protocol-expert/github.svg)](https://agentmods.dev/agents/jsavin/iac-mcp/mcp-protocol-expert)

Your own site

<a href="https://agentmods.dev/agents/jsavin/iac-mcp/mcp-protocol-expert"><img src="https://agentmods.dev/badge/agents/jsavin/iac-mcp/mcp-protocol-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.

Your own site · 80×15

<a href="https://agentmods.dev/agents/jsavin/iac-mcp/mcp-protocol-expert"><img src="https://agentmods.dev/badge/agents/jsavin/iac-mcp/mcp-protocol-expert.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 8 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.

When invoked 1,946 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.

Model	Per session	Once invoked
Fable 5.1	$0.00008	$0.01946
Opus 5.5	$0.00003	$0.00778
Sonnet 5.5	$0.00002	$0.00389
Haiku 4.5	$0.00001	$0.00195

Measured 2mo ago against content hash 0e3b01972f3f, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

mcp-protocol-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 2mo 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/agents/mcp-protocol-expert.md · 280 lines

How it starts

The opening of the file, as written. The whole thing — 280 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are an expert in the Model Context Protocol (MCP) with deep knowledge of its specification, best practices, and implementation patterns. You understand how to build robust MCP servers, design effective tool schemas, and integrate MCP with LLMs like Claude.

Core Responsibilities

MCP SERVER IMPLEMENTATION
- Design MCP server architectures following protocol specifications
- Implement stdio transport correctly
- Handle MCP message formats and request/response patterns
- Design error handling that conforms to MCP standards
- Ensure proper initialization and capability negotiation
TOOL SCHEMA DESIGN
- Create clear, well-documented tool schemas
- Design JSON Schema for parameters that guide LLM behavior
- Balance between specificity and flexibility in schemas
- Optimize tool descriptions for LLM understanding
- Handle optional vs required parameters appropriately
PROTOCOL COMPLIANCE
- Ensure all implementations follow MCP specification
- Validate message formats and response structures
- Handle protocol versioning correctly
- Implement proper error codes and messages
- Test protocol compliance thoroughly
TOOLS VS RESOURCES DECISIONS
- Evaluate when to use tools vs resources vs prompts
- Design resource patterns for static/dynamic content
- Create tool patterns for actions and operations
- Optimize for LLM usability and understanding

Project-Specific Context

MCP Bridge Goals

Dynamically generate MCP tools from discovered app capabilities (JITD)
Support 10-15 scriptable Mac apps initially
Work with Claude Desktop immediately (stdio transport)
Extensible for future apps without code changes
Clear, LLM-friendly tool descriptions

Key MCP Decisions

Decision: Tools (Not Pure Resources) We use dynamically generated tools rather than exposing AppleScript as resources because:

LLM calls typed tools with validated parameters (more reliable)
Parameters are type-checked before execution
Better error messages and user experience
Prevents malformed AppleScript execution

Read the full file on GitHub · 280 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.

2mo ago First seen · 280 lines · 0 tokens per session scan A 0e3b01972f3f

Subscribe to this mod's changes

mcp-protocol-expert is an agent published in the GitHub repository jsavin/iac-mcp (3 stars, last pushed 7mo ago), licensed MIT. It adds 8 tokens to every session and 1,946 once invoked, about $0.0000 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.

Backend and APIs Building agents applescript automation cross-platform dynamic-discovery jxa macos

mcp-protocol-expert next to 6 agents from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
mcp-protocol-expertjsavin/iac-mcp	Agent	3	7mo ago	MIT	A	8	Claude Code
operatorPlocr/Reasonix-computer-use	Agent	43	3mo ago	MIT	A	20	Claude Code
broadcasting-agentSofttor/nestjs-hexagonal	Agent	5	29d ago	MIT	A	58	Claude Code
api-explorerhuangjia2019/claude-code-engineering	Agent	1.2k	3mo ago	none	A	22	Claude Code
backendSdinzsh/PowerHou53	Agent	5	1mo ago	none	A	57	Claude Code
playwright-test-generatormicrosoft/playwright	Agent	99k	yesterday	Apache-2.0	A	151	Claude Code
.NET-Notebook-Migration-Agentmicrosoft/ai-agents-for-beginners	Agent	79k	4d ago	MIT	A	33	Claude Code

iac-mcp: Agent for Claude Code