mcp-server-review

mcp-server-review is a skill for Claude Code, Codex from pnp/copilot-prompts. It costs 93 tokens per session (2,803 once invoked), scanned A, original, MIT.

A compliance reviewer for MCP servers. MCP, or Model Context Protocol, is a standard way for AI applications to communicate with tools and data servers.

In plain words
What is it for?
Use it to review, audit, upgrade, or migrate an MCP server. It covers Streamable HTTP and standard input/output transports and can consider compatibility with older MCP clients.
Why use it?
It identifies code that does not match the 2026-07-28 MCP revision, including obsolete connection behavior, missing requirements, and security issues. It also separates confirmed findings from things the available code cannot establish.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one.

Good fit Use it to review, audit, upgrade, or migrate an MCP server. It covers Streamable HTTP and standard input/output transports and can consider compatibility with older MCP clients.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/pnp/copilot-prompts/mcp-server-review
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.

Any agent
npx skills add pnp/copilot-prompts --skill mcp-server-review
Clone the repo
git clone --depth 1 https://github.com/pnp/copilot-prompts

Made for: Claude Code, Codex.

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-server-review

README.md
[![agentmods](https://agentmods.dev/badge/skills/pnp/copilot-prompts/mcp-server-review/github.svg)](https://agentmods.dev/skills/pnp/copilot-prompts/mcp-server-review)
Your own site
<a href="https://agentmods.dev/skills/pnp/copilot-prompts/mcp-server-review"><img src="https://agentmods.dev/badge/skills/pnp/copilot-prompts/mcp-server-review/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 mcp-server-review

Your own site · 80×15
<a href="https://agentmods.dev/skills/pnp/copilot-prompts/mcp-server-review"><img src="https://agentmods.dev/badge/skills/pnp/copilot-prompts/mcp-server-review.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 93 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,803 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. Third-party audits
  • NVIDIA SkillSpector pass 7 Sept 2026
How audits are shown
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 $0.00093 $0.02803
Opus 5 $0.00046 $0.01401
Sonnet 5 $0.00019 $0.00561
Haiku 4.5 $0.00009 $0.00280

Measured 10d ago against content hash 2f7e4fc7cfe4, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

mcp-server-review 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 10d 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.

samples/skills/mcp-server-review/SKILL.md · 180 lines

How it starts

The opening of the file, as written. The whole thing — 180 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Review an MCP Server Against the 2026-07-28 Revision

Protocol revision 2026-07-28 is the largest change to the Model Context Protocol since it launched. Server code written before it still compiles, still passes its own tests, and still looks correct in review. It simply speaks a shape that modern clients no longer send.

This skill finds that gap. It reviews an MCP server implementation against the current revision, reports what is obsolete, what is missing, and what is a security requirement rather than a nicety, and it distinguishes between a genuine finding and something it could not determine from the code available.

Before Starting

Critical: Always ask the user for the following information before proceeding:

  1. Server code - the repository, folder, or files implementing the MCP server
  2. Target revision - the protocol revision the server intends to implement. Defaults to 2026-07-28 if not specified
  3. Transport - Streamable HTTP or stdio. Defaults to Streamable HTTP if not specified, since the header rules below apply only to it
  4. Backward compatibility requirement - whether the server must still serve clients on 2025-03-26 through 2025-11-25. Defaults to no

If the user doesn't provide all details upfront, ask for the missing ones before proceeding.

Do not infer the revision from an SDK version number. SDK majors and protocol revisions move independently. Read what the code actually sends and accepts.

Output Structure

Produce a review report with these sections:

  1. Revision verdict - one line stating which revision the code appears to implement, and the evidence for that conclusion
  2. Blocking findings - table of Finding | Location | Why it breaks | Fix
  3. Security findings - the spec's security requirements with their MUST or SHOULD levels, listed separately because they are not stylistic
  4. Deprecated but working - features on the twelve month removal clock
  5. Not determined - checks that could not be evaluated from the code provided, and what to look at to resolve each
  6. Nothing to change - the checks that passed

Read the full file on GitHub · 180 lines

Files

What ships with it

3 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

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. 10d ago First seen · 180 lines · 93 tokens per session scan A 2f7e4fc7cfe4

Subscribe to this mod's changes

mcp-server-review is a skill published in the GitHub repository pnp/copilot-prompts (875 stars, last pushed 2d ago), licensed MIT. It adds 93 tokens to every session and 2,803 once invoked, about $0.0005 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 2026-08-30.

Related

Other skills, from other repositories