flipperzero-mcp CLAUDE.md

flipperzero-mcp CLAUDE.md is an instructions file for coding agents from millsymills-com/flipperzero-mcp. It costs 837 tokens per session, scanned A, original, MIT.

Repository instructions for an MCP server that connects a Flipper Zero device to coding tools over USB or Wi‑Fi. MCP is a standard way for an AI assistant to call tools provided by another program.

In plain words
What is it for?
Use them when developing or reviewing the flipperzero-mcp repository. They cover project-specific behavior, protobuf version requirements, tool idempotency, and self-auditing.
Why use it?
They document the project's required standards and important communication rules, such as keeping logs separate from the JSON-RPC tool channel. They also provide a command for checking the implementation against those standards.

Instructions file

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 instructions/millsymills-com/flipperzero-mcp/claude-md
Clone the repo
git clone --depth 1 https://github.com/millsymills-com/flipperzero-mcp

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 flipperzero-mcp CLAUDE.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/millsymills-com/flipperzero-mcp/claude-md.svg)](https://agentmods.dev/instructions/millsymills-com/flipperzero-mcp/claude-md)
Your own site
<a href="https://agentmods.dev/instructions/millsymills-com/flipperzero-mcp/claude-md"><img src="https://agentmods.dev/badge/instructions/millsymills-com/flipperzero-mcp/claude-md.svg" alt="Measured on agentmods" height="20"></a>
Per session 837 This file is loaded in full into every session.
When invoked 837 The same file — it is already loaded in full.
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 $0.00837 $0.00837
Opus 5 $0.00418 $0.00418
Sonnet 5 $0.00167 $0.00167
Haiku 4.5 $0.00084 $0.00084

Measured 4d ago against content hash 98fb5c9dfab0, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

flipperzero-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 4d 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 · 75 lines

How it starts

The opening of the file, as written. The whole thing — 75 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 repository.

Inherits the workspace conventions in ../CLAUDE.md and the global standards in ~/.claude/CLAUDE.md. Defer to those for code-quality limits, tooling, and workflow.

Canonical standards

This server is graded against the canonical MCP standards at consistency-check/docs/standards/ (mcp.md + python.md + mcp-protocol.md + stages.md). Self-audit before opening a PR:

uv run consistency-check audit --repo flipperzero-mcp

Stage is declared in the README ## Status section (currently S2); the auditor scopes rules to the declared stage.

Project

stdio MCP server (FastMCP) for the Flipper Zero, speaking protobuf RPC over USB or WiFi.

Project specifics

  • stdout is the JSON-RPC channel. All logging goes to stderr (see _logging.py). Never print to stdout.
  • Tool idempotency annotations are behavioral, not just end-state based. Tools that cycle state on every call (for example disconnect/reconnect) must use idempotentHint=False even if they usually converge to the same final state.
  • Protobuf runtime is pinned to protobuf==6.33.5 (forward-compatible with the vendored bindings in src/flipperzero_mcp/rpc/protobuf_gen/, which embed a 6.33.2 assertion; bumped for CVE-2026-0994). Keep the pin's major.minor at 6.33. Regenerate the bindings if you change the pin's major.minor or the .proto sources in proto/.
  • Harvest provenance. The transport layer (src/flipperzero_mcp/transport/), the protobuf RPC implementation (src/flipperzero_mcp/rpc/protobuf_rpc.py and proto/), and the ESP32 firmware (firmware/tcp_uart_bridge/) are harvested from busse/flipperzero-mcp under the MIT License. See NOTICE.
  • Integration tests require a real Flipper. They are marked integration (plus usb/wifi) and are local-only. The default suite is uv run pytest -m "not integration".
  • Firmware is not built in CI. It is harvested as-is for WiFi users.
  • firmware/ module (flavor.py, bundles.py, installer.py, codes.py): classifies firmware flavor (Official vs Momentum), resolves and downloads bundles with sha256 verification, and drives the on-device installer via chunked storage_write (required for large files). Gated behind FLIPPER_ENABLE_FIRMWARE_FLASH=true (plus FLIPPER_ENABLE_WRITE_TOOLS=true).
  • Dual-firmware golden fixtures. Recorded fixtures live under tests/golden/fixtures/{momentum,official}/. Re-record after changing the protobuf schema or adding tools that have firmware-specific output.

Read the full file on GitHub · 75 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. 4d ago First seen · 75 lines · 837 tokens per session scan A 98fb5c9dfab0

Subscribe to this mod's changes

flipperzero-mcp CLAUDE.md is an instructions file published in the GitHub repository millsymills-com/flipperzero-mcp (4 stars, last pushed 6d ago), licensed MIT. It adds 837 tokens to every session, about $0.0042 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.