mcp-review

mcp-review is a skill for Claude Code from stacklok/toolhive-catalog. It costs 51 tokens per session (3,140 once invoked), scanned A, original, Apache-2.0.

A review guide for checking MCP server registry submissions and changes. MCP servers are services that let AI tools use external tools or data.

In plain words
What is it for?
Review server.json files, pull requests, and proposed MCP server updates against the registry’s field and inclusion rules.
Why use it?
It helps catch specification errors, security issues, incomplete entries, and submissions placed in the wrong registry before they are accepted.

Skill for Claude Code

Written for Claude Code: allowed-tools in frontmatter.

Good fit Review server.json files, pull requests, and proposed MCP server updates against the registry’s field and inclusion rules.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/stacklok/toolhive-catalog/mcp-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 stacklok/toolhive-catalog --skill mcp-review
Clone the repo
git clone --depth 1 https://github.com/stacklok/toolhive-catalog

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 mcp-review

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

Your own site · 80×15
<a href="https://agentmods.dev/skills/stacklok/toolhive-catalog/mcp-review"><img src="https://agentmods.dev/badge/skills/stacklok/toolhive-catalog/mcp-review.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 51 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 3,140 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 1 finding. A grade says what 26 rules found in the file — not that it is safe. Third-party audits
  • NVIDIA SkillSpector warn 7 Sept 2026
SkillSpector: 1 finding, up to medium

These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →

  • medium Data Exfiltration · line 58
    Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
    Fix: Verify the destination URL is trusted and necessary. Remove or replace with documented APIs. Ensure no secrets, tokens, or PII are transmitted.
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 $0.00051 $0.03140
Opus 5.5 $0.00020 $0.01256
Sonnet 5.5 $0.00010 $0.00628
Haiku 4.5 $0.00005 $0.00314

Measured 1mo ago against content hash f218f119daad, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

mcp-review scanned grade A with 1 finding 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.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

curl -sS -X POST "<remote-url>" \
.claude/skills/mcp-review/SKILL.md · 248 lines

How it starts

The opening of the file, as written. The whole thing — 248 lines — stays where its author put it; the contents beside it link to each section on GitHub.

MCP Server Specification Review

You are an expert reviewer for the ToolHive Registry. Evaluate server.json files and MCP server submissions for spec compliance, security, registry inclusion criteria, and completeness.

For detailed field specs, see server-json-spec.md. For full registry inclusion criteria, see registry-criteria.md and server-criteria.md.

Registry Layout (read first)

The repo has two server registries, and submissions land in the wrong one often enough to check first:

  • registries/toolhive/servers/ — the contribution path. All new submissions go here.
  • registries/official/ — a curated subset maintained by the ToolHive team. Contributors should not hand-add entries here.

Both normalize every entry under the io.github.stacklok publisher namespace (name: io.github.stacklok/<server-name> and the _meta publisher key), with a repo-hosted icon.svg. A PR that adds a file under registries/official/, or uses a third-party namespace like io.github.<contributor>, is a finding — not a style nit. task catalog:validate does not catch either; verify by eye and by comparing against a neighboring entry (jq '{name, ns: (._meta[...]|keys)}' <neighbor>/server.json).

Review Workflow

Step 1: Identify Change Scope

Determine what you're reviewing:

  • New server submission → Full review (spec + repository assessment + inclusion criteria)
  • Version update → Focused review (changed fields, both identifier and _meta key updated, changelog)
  • Config change → Targeted review (just the changed aspects + security implications)

Step 2: Validate server.json

Read the server.json and check, in order:

  1. Location & namespace — file is under registries/toolhive/servers/<name>/; name and the _meta publisher key are both io.github.stacklok (see Registry Layout)
  2. Server type — packages (container) or remotes (remote)?
  3. Required top-level fields — $schema, name, description, title, version, repository, icons
  4. Package/remote config — identifier format, transport type valid for server type
  5. Extension key match — _meta key must exactly equal packages[0].identifier or remotes[0].url
  6. Extension fields — tier, status, tools, overview all present and valid
  7. Tier matches maintainer — Official is reserved for the ToolHive team, the MCP spec authors, or the platform owner of the integrated service. A third-party contributor's server is Community, regardless of what the PR claims (don't trust the field — check who maintains the repo)
  8. Icons — repo-hosted icon.svg (mimeType: image/svg+xml) AND an icon.svg file present in the server dir. Flag external icon URLs (e.g. a vendor PNG) — they're a stability/supply-chain risk and break the convention
  9. Overview format — starts with ## Title\n\n followed by 3-5 sentences
  10. No auto-populated fields — reject if metadata.* or tool_definitions present in new submissions
  11. Remote-specific — tags include "remote"; oauth_config present if OAuth required
  12. Container-specific — transport.url present when type is streamable-http

Read the full file on GitHub · 248 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. 1mo ago First seen · 248 lines · 51 tokens per session scan A f218f119daad

Subscribe to this mod's changes

mcp-review is a skill published in the GitHub repository stacklok/toolhive-catalog (24 stars, last pushed yesterday), licensed Apache-2.0. It adds 51 tokens to every session and 3,140 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it A with 1 finding (makes network calls). 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

code-review

Review code for quality, maintainability, and correctness.

PrefectHQ/fastmcp · 10 tokens

release-pr-review

Review pass on an open release PR (release/ → main) — the step between git-wrapup and release-and-publish when a project releases in gated release PR mode. Reads the PR's commit range through the code-simplifier lens plus a correctness review, verifies whatever an automated reviewer left on the PR, lands fixes as…

cyanheads/obsidian-mcp-server · 136 tokens

code-simplifier

Cleanup pass that edits the working tree — over a session's uncommitted changes, or a named path or whole codebase. Reads git diff (or the named target) and simplifies, consolidates, and aligns code with the existing codebase — modernize syntax, cut unnecessary complexity and slop, consolidate duplicated logic, catch…

cyanheads/obsidian-mcp-server · 160 tokens

code-review

Reviews a supplied code path or diff for correctness, security, maintainability, and style without executing or modifying it.

sandbaseai/sandbase-harness · 25 tokens

github-commenting

How to post clean, rich, deduplicated GitHub PR review comments — suggestion blocks, multi-line anchors, markers, formatting rules. Load before posting or fixing any PR comment.

juspay/neurolink · 41 tokens

code-review

Review ServiceNow server-side scripts for ES5 violations, ACL/injection/XSS issues, N+1 queries, missing setLimit/error handling, hard-coded sysids, and business-rule recursion risks.

serac-labs/serac · 43 tokens