---
id: cobs
version: "1.2.2"
license: ---------------------------------------------------------------------------- Copyright (c) 2010 Craig McQueen Permission is hereby granted, free of charge, to any person obtaining a copy of this… (full text in the JSON record)
license_treatment: permissive
maintenance: aging
---

# cobs — Consistent Overhead Byte Stuffing (COBS)

License: permissive · Maintenance: aging · Downloads: 119.2K/mo

## What it is and what it does

COBS is a byte-stuffing algorithm that encodes arbitrary binary data into a form containing no zero bytes (0x00), enabling zero bytes to serve as reliable packet delimiters in serial and network protocols. The package provides both basic COBS and COBS/R (a variant that reduces overhead) via Python modules with encode and decode functions. It includes a C extension for performance and a pure Python fallback for portability.

The encoding adds predictable overhead: always 1 byte plus potentially n/254 additional bytes for longer packets. This is more efficient than escape-based methods like PPP, which can double packet size in worst cases. The package accepts any buffer-protocol-compatible input (bytes, bytearray) and returns encoded/decoded bytes; framing (adding or stripping delimiter bytes) is left to the application.

Use it for:

- Implement packet framing in serial communication protocols where zero bytes mark packet boundaries.
- Encode binary data for transmission over channels that cannot safely transmit zero bytes.
- Reduce packet size overhead compared to escape-based byte stuffing in bandwidth-constrained protocols.
- Use COBS/R variant to minimize encoding overhead for small message payloads in embedded systems.

## Worth the install?

AI-flagged interpretation of the facts above — verify before relying.

Encodes and decodes data using Consistent Overhead Byte Stuffing (COBS), a method that eliminates zero bytes from encoded packets to enable reliable packet framing in communication protocols.

Yes. COBS is a well-established, stable algorithm with a mature implementation (Production/Stable, MIT license). Install it if you need reliable packet framing in serial/network protocols or must eliminate zero bytes from binary data. Medium install friction is acceptable given the C extension fallback. No known vulnerabilities.

## Install

pip install cobs
uv add cobs
poetry add cobs

## Installing cobs

Before you install: Medium install friction due to C extension compilation; pure Python fallback available. Package is stable (Production/Stable status) and actively maintained with recent release (2025-07-20), though maintenance cadence is aging.

License in practice: MIT license (permissive) allows unrestricted use, modification, and distribution with minimal restrictions—suitable for both open-source and proprietary projects.

Quickstart:

pip install cobs

from cobs import cobs
encoded = cobs.encode(b'Hello world\x00This is a test')
decoded = cobs.decode(encoded)

Requires Python >= 3.6; C extension compilation may fail on some platforms, but pure Python implementation is available as fallback.

Verify before relying:

- Performance characteristics of C extension vs. pure Python implementation for typical packet sizes.
- Whether the package handles edge cases like empty input or maximum-length packets as documented.

## Package facts

- License: ---------------------------------------------------------------------------- Copyright (c) 2010 Craig McQueen Permission is hereby granted, free of charge, to any person obtaining a copy of this… (full text in the JSON record) (permissive)
- Python support: supports_current
- Install friction: medium
- Maintenance: aging
- Downloads: 119.2K/month (top 15,000 on PyPI)
- Known vulnerabilities: none known

## Tags

byte stuffing encoding, cobs packet framing, zero byte elimination, serial protocol encoding, packet boundary delimiter, communication protocol encoding, cobs/r reduced overhead, packet-framing, serial-protocol, byte-encoding

[View on SkillFed](https://skillfed.io/packages/cobs) · [View on PyPI](https://pypi.org/project/cobs/)