verify-ps-mcp

verify-ps-mcp is a skill for Claude Code, Codex, Cursor from planetscale/mcp-server. It costs 56 tokens per session (1,480 once invoked), scanned A, original, Apache-2.0.

A verification workflow for a PlanetScale MCP server, where MCP is a standard way for an agent to call tools through a server.

In plain words
What is it for?
Starting the server over standard input and output, listing tools, calling one, and recording proof of its response before pushing changes.
Why use it?
It checks the real server and a real tool call, instead of relying only on linting or type checks.

Skill for Claude CodeCodexCursor

About the project

planetscale/mcp-server is an open-source collection of TypeScript tools and shared code for the PlanetScale MCP server, which lets compatible coding agents interact with PlanetScale. It supports local development, testing, and deployment of the repository’s maintained tools; other server features are generated from PlanetScale’s API specification. The catalogue entries provide instructions, an MCP configuration, and a skill for working with it.

planetscale/mcp-server · 4 stars · on GitHub

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 skills/planetscale/mcp-server/verify-ps-mcp
Any agent
npx skills add planetscale/mcp-server --skill verify-ps-mcp
Clone the repo
git clone --depth 1 https://github.com/planetscale/mcp-server

Made for: Claude Code, Codex, Cursor.

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 verify-ps-mcp

README.md
[![agentmods](https://agentmods.dev/badge/skills/planetscale/mcp-server/verify-ps-mcp.svg)](https://agentmods.dev/skills/planetscale/mcp-server/verify-ps-mcp)
Your own site
<a href="https://agentmods.dev/skills/planetscale/mcp-server/verify-ps-mcp"><img src="https://agentmods.dev/badge/skills/planetscale/mcp-server/verify-ps-mcp.svg" alt="Measured on agentmods" height="20"></a>
Per session 56 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,480 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 $0.00056 $0.01480
Opus 5 $0.00028 $0.00740
Sonnet 5 $0.00011 $0.00296
Haiku 4.5 $0.00006 $0.00148

Measured 5d ago against content hash 7e2efb0ee0d0, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

verify-ps-mcp 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 5d ago.

The scan reads SKILL.md. This mod also ships 1 executable file (drive.mjs), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

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.

.cursor/skills/verify-ps-mcp/SKILL.md · 142 lines

How it starts

The opening of the file, as written. The whole thing — 142 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Verify the PlanetScale MCP server

The user-facing surface of this repo is an MCP server spoken over stdio: a client (Cursor, MCP Inspector, the hosted Gram deployment) starts src/server.ts, lists tools, and calls them. Verification means being that client — start the real server, call a real tool the way a real client would, and keep the response.

npm run lint only proves the code typechecks. It does not prove a tool returns the right fields, so it is never sufficient as a proof on its own.

Launch

There is no long-lived server to babysit. Each drive spawns its own node_modules/.bin/tsx src/server.ts child over stdio and kills it when the run ends, so runs are isolated from each other and can overlap safely.

One-time setup in a fresh checkout:

npm install
cp .env.example .env   # then fill in PLANETSCALE_API_TOKEN

The server loads .env itself through dotenv/config. Auth resolves in src/lib/auth.ts: PLANETSCALE_OAUTH2_ACCESS_TOKEN wins if set, otherwise PLANETSCALE_API_TOKEN (a <token-id>:pscale_tkn_... service token) is sent as-is. Never echo either value into a terminal, a log, or an evidence file.

Doctor

Run this first whenever anything looks wrong. It is read-only: it starts a server, initializes the MCP session, lists tools, and checks the thirteen expected tools are registered.

node .cursor/skills/verify-ps-mcp/drive.mjs doctor

Healthy output names all thirteen tools and exits 0:

OK 13 tools registered: execute_read_query, execute_write_query, get_insights, get_payment_method_setup, get_postgres_logs, get_query_tag, list_cluster_sizes, list_query_error_executions, list_query_error_patterns, list_query_tag_summaries, list_query_tags, search_documentation, update_payment_method

A crash here means the server itself is broken (a bad import, a Zod schema that throws at construction), not the tool you were investigating. Fix that before driving a feature.

Drive

node .cursor/skills/verify-ps-mcp/drive.mjs call <tool> '<json args>' \
  [--expect <substring>] [--label <name>]

Read the full file on GitHub · 142 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. 5d ago First seen · 142 lines · 56 tokens per session scan A 7e2efb0ee0d0

Subscribe to this mod's changes

verify-ps-mcp is a skill published in the GitHub repository planetscale/mcp-server (4 stars, last pushed yesterday), licensed Apache-2.0. It adds 56 tokens to every session and 1,480 once invoked, about $0.0003 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.