nanobanana-mcp: Instructions file for Claude Code

CLAUDE.md

nanobanana-mcp CLAUDE.md is an instructions file for Claude Code from pistachiomatt/nanobanana-mcp. It costs 935 tokens per session, scanned A, original, MIT.

A set of Claude Code instructions for the NanoBanana MCP project, a server that exposes Google's Gemini image generation, image editing, and vision features. It documents how the project is built, run, and organized.

In plain words
What is it for?
Use it when changing or running this TypeScript MCP server, including rebuilding its committed output after source changes. It also documents command installation for Claude Code and Cursor.
Why use it?
It keeps development work consistent by recording the build commands, architecture, session handling, and tool behavior in one place.

Instructions file for Claude Code

Written for Claude Code: the file is CLAUDE.md. Also seen: mentions CLAUDE.md; mentions Claude Code.

This is pistachiomatt/nanobanana-mcp's own configuration. It tells Claude Code how to work on nanobanana-mcp itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything nanobanana-mcp configures →

Reuse

Borrowing it

Nothing to install: this file belongs to pistachiomatt/nanobanana-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/pistachiomatt/nanobanana-mcp/main/CLAUDE.md
Clone the repo
git clone --depth 1 https://github.com/pistachiomatt/nanobanana-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.

agentmods badge for nanobanana-mcp CLAUDE.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/pistachiomatt/nanobanana-mcp/claude-md/github.svg)](https://agentmods.dev/instructions/pistachiomatt/nanobanana-mcp/claude-md)
Your own site
<a href="https://agentmods.dev/instructions/pistachiomatt/nanobanana-mcp/claude-md"><img src="https://agentmods.dev/badge/instructions/pistachiomatt/nanobanana-mcp/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.

agentmods 80×15 button for nanobanana-mcp CLAUDE.md

Your own site · 80×15
<a href="https://agentmods.dev/instructions/pistachiomatt/nanobanana-mcp/claude-md"><img src="https://agentmods.dev/badge/instructions/pistachiomatt/nanobanana-mcp/claude-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 935 This file is loaded in full into every session.
When invoked 935 The same file — it is already loaded in full.
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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00935 $0.00935
Opus 5.5 $0.00374 $0.00374
Sonnet 5.5 $0.00187 $0.00187
Haiku 4.5 $0.00093 $0.00093

Measured 2mo ago against content hash 93901991d2fd, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

nanobanana-mcp CLAUDE.md 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.md · 85 lines

How it starts

The opening of the file, as written. The whole thing — 85 lines — stays where its author put it; the contents beside it link to each section on GitHub.

CLAUDE.md

Guidance for Claude Code when working in this repo.

What This Is

NanoBanana MCP — a Model Context Protocol server exposing Google Gemini's image generation, editing, and vision/chat capabilities over stdio. Single-file server at src/index.ts.

Build & Run

npm run build        # tsc → dist/
npm run dev          # tsx watch (hot reload)
npm run start        # node dist/index.js

dist/ is committed. After any change to src/index.ts, always npm run build so dist/index.js stays in sync. This enables direct GitHub-based npx usage without a build step.

Architecture

Everything lives in src/index.ts. No multi-file abstractions — it's intentionally a single file. The structure:

  1. CLI handler (lines ~7–47) — --install-commands <claude-code|cursor> copies slash commands
  2. Gemini API layer — callGeminiImageAPI() does raw REST calls with streaming response parsing
  3. Session management — conversations Map keyed by conversation_id, holding chat history, image history, aspect ratio, and model selection
  4. MCP tools — registered via ListToolsRequestSchema / CallToolRequestSchema

Tools

Tool Purpose
gemini_chat Multi-turn conversation with optional images
gemini_generate_image Image generation with consistency/reference support
gemini_edit_image Natural language image editing
set_aspect_ratio Set session aspect ratio (required before image gen/edit)
set_model Switch flash/pro per session
get_image_history List generated/edited images in session
clear_conversation Reset session context
batch Execute multiple tools in parallel, return all results

Prompt Resolution

All three content tools (gemini_chat, gemini_generate_image, gemini_edit_image) support file-based prompts as an alternative to inline strings:

  • message / message_file
  • prompt / prompt_file
  • edit_prompt / edit_prompt_file

Read the full file on GitHub · 85 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. 2mo ago First seen · 85 lines · 935 tokens per session scan A 93901991d2fd

Subscribe to this mod's changes

nanobanana-mcp CLAUDE.md is an instructions file published in the GitHub repository pistachiomatt/nanobanana-mcp (0 stars, last pushed 7mo ago), licensed MIT. It adds 935 tokens to every session, about $0.0037 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.

Related

nanobanana-mcp CLAUDE.md compared with instructions from other repositories

nanobanana-mcp CLAUDE.md next to 6 instructions from other repositories
ModTypeStarsLast pushLicenceScanTokensWorks with
nanobanana-mcp CLAUDE.mdpistachiomatt/nanobanana-mcpInstructions file07mo agoMITA935Claude Code
vscode buildNext.instructions.mdmicrosoft/vscodeInstructions file196ktodayMITA9,719GitHub Copilot
next.js AGENTS.mdvercel/next.jsInstructions file144ktodayMITA8,130Codex, OpenCode
vscode oss-third-party-notices.instructions.mdmicrosoft/vscodeInstructions file196ktodayMITA5,685GitHub Copilot
langchain AGENTS.mdlangchain-ai/langchainInstructions file150ktodayMITA4,680Codex, OpenCode
deepseek-harness AGENTS.mddeepseek-ai/deepseek-harnessInstructions file283ktodayMITA4,342Codex, OpenCode
vscode copilot-instructions.mdmicrosoft/vscodeInstructions file196ktodayMITA3,153GitHub Copilot