What is mcp-server CLAUDE.md?

A repository-specific instruction file for Claude and other coding assistants, explaining the project structure, file conventions, verification steps, and commit guidance.

Use it when working on the mcp-server project, especially when editing its server, configuration, scripts, or runtime files.

It gives an assistant the local rules it needs to make changes that fit the repository and to verify different kinds of files correctly.

How LeszczynskiKarol/mcp-server configures Claude Code

Reuse

Borrowing it

Nothing to install: this file belongs to LeszczynskiKarol/mcp-server. 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/LeszczynskiKarol/mcp-server/main/CLAUDE.md

Clone the repo

git clone --depth 1 https://github.com/LeszczynskiKarol/mcp-server

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/instructions/leszczynskikarol/mcp-server/claude-md/github.svg)](https://agentmods.dev/instructions/leszczynskikarol/mcp-server/claude-md)

Your own site

<a href="https://agentmods.dev/instructions/leszczynskikarol/mcp-server/claude-md"><img src="https://agentmods.dev/badge/instructions/leszczynskikarol/mcp-server/claude-md/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/instructions/leszczynskikarol/mcp-server/claude-md"><img src="https://agentmods.dev/badge/instructions/leszczynskikarol/mcp-server/claude-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 4,100 This file is loaded in full into every session.

When invoked 4,100 The same file — it is already loaded in full.

Security scan C 1 finding. 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	$0.04100	$0.04100
Opus 5.5	$0.01640	$0.01640
Sonnet 5.5	$0.00820	$0.00820
Haiku 4.5	$0.00410	$0.00410

Measured 2mo ago against content hash 124340175f39, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade C, and why

mcp-server CLAUDE.md scanned grade C 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 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.

Reaches for credential fileshighPrivilege escalation

SSH keys, cloud credentials, git-credentials, .npmrc, /etc/shadow: reading these is how a config file becomes a credential leak.

The local AWS CLI profile (from `~/.aws/credentials` or env vars) must have

CLAUDE.md · 317 lines

How it starts

The opening of the file, as written. The whole thing — 317 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Instructions for Claude when working on this repo

This file is read by Claude (and other AI assistants) at the start of each session. The CORE rules (Trust the write, sandbox isolation, anti-loop, artifacts vs MCP, file-tool hierarchy, Windows path escaping, chunking long outputs) live in my Anthropic user preferences and apply to ALL repos. See CLAUDE_PREFERENCES.md in this repo for the full text.

This file adds context specific to mcp-server.

Project layout

server.js — main MCP server (single file by design, ~1700 lines after hardening). Don't split into multiple files unless I explicitly ask.
hosts.json — gitignored, server/key registry for ssh_exec, postgres_query, pm2_status.
.env — gitignored, OAuth password + base URL + GitHub token.
oauth-state.json — gitignored, runtime persistence of OAuth clients and tokens (tokens stored sha256-hashed, never plaintext).
known_hosts — gitignored, SSH host fingerprints (pinned on first connect).
logs/mcp.log — gitignored.
.pid — gitignored, server.js writes its own PID here on boot. Used by restart-mcp.ps1 to find the exact node.exe to kill.
restart-mcp.ps1 — selective restart (see Restart workflow below).
watchdog.ps1 — runs every 2 minutes via Task Scheduler "MCP Server Watchdog", respawns the server if both the node.exe AND its start-mcp.bat loop are dead. Logs to logs/watchdog.log.
tmp/ — gitignored, designated scratch directory. Anything written here is treated as disposable and gets pruned by cleanup-tmp.ps1 (see Scratch files convention below).
cleanup-tmp.ps1 — Task Scheduler "MCP Tmp Cleanup" (hourly), universal scratch cleanup. In one pass:
1. Deletes ANY file in tmp/ older than -MaxAgeHours (default 24), regardless of name / extension. Recursive.
2. Sweeps D:\ root for files matching scratch patterns (tmp_*, temp_*, _tmp*, _temp*, scratch_*, claude_*, mcp_*, draft_*, out_*.txt, output_*.txt, chunk_*.txt, _part_*, test_utf8.*, test_*.tmp, *.tmp, *.bak, *.old, *~, *.scratch) older than -RootMaxAgeHours (default 24). Flags: -DryRun, -SkipRoot, -MaxAgeHours N, -RootMaxAgeHours N. Add new patterns to $RootPatterns in the script when needed; protect specific real files via $RootExclude. Logs to logs/cleanup-tmp.log.

Read the full file on GitHub · 317 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 · 317 lines · 4,100 tokens per session scan C 124340175f39

Subscribe to this mod's changes

mcp-server CLAUDE.md is an instructions file published in the GitHub repository LeszczynskiKarol/mcp-server (2 stars, last pushed 2mo ago), licensed MIT. It adds 4,100 tokens to every session, about $0.0164 per session on Opus 5.5. A static security scan graded it C with 1 finding (reaches for credential files). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

anthropic aws claude claude-ai developer-tools express

mcp-server CLAUDE.md next to 6 instructions from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
mcp-server CLAUDE.mdLeszczynskiKarol/mcp-server	Instructions file	2	2mo ago	MIT	C	4,100	Claude Code
vscode buildNext.instructions.mdmicrosoft/vscode	Instructions file	195k	yesterday	MIT	A	9,719	GitHub Copilot
next.js AGENTS.mdvercel/next.js	Instructions file	143k	yesterday	MIT	A	7,461	Codex, OpenCode
langchain AGENTS.mdlangchain-ai/langchain	Instructions file	149k	yesterday	MIT	A	4,680	Codex, OpenCode
vscode oss-third-party-notices.instructions.mdmicrosoft/vscode	Instructions file	195k	yesterday	MIT	A	5,516	GitHub Copilot
vscode copilot-instructions.mdmicrosoft/vscode	Instructions file	195k	yesterday	MIT	A	3,093	GitHub Copilot
deepseek-harness AGENTS.mddeepseek-ai/deepseek-harness	Instructions file	274k	2d ago	MIT	A	4,313	Codex, OpenCode

mcp-server: Instructions file for Claude Code