Skip to main content
CVE Vulnerability Database
Vulnerability Database/CVE-2026-77087

CVE-2026-77087: Paperclip RCE Vulnerability via DNS Rebinding

CVE-2026-77087 is a remote code execution flaw in Paperclip before version 0.3.1 that exploits DNS rebinding to execute arbitrary commands. This post covers the technical details, affected versions, security impact, and mitigation.

Published:

CVE-2026-77087 Overview

CVE-2026-77087 is a remote code execution vulnerability in Paperclip versions before 0.3.1. The flaw exists in the default local_trusted mode, which fails to validate HTTP Host headers on inbound requests. An attacker can craft a malicious webpage that leverages DNS rebinding to reach the Paperclip service running on a developer's local machine. Once the browser is rebound to the local interface, the attacker issues authenticated API calls that reach the process adapter, resulting in arbitrary command execution. The weakness is categorized as [CWE-862] Missing Authorization.

Critical Impact

Attackers can execute arbitrary operating system commands on any developer machine running vulnerable Paperclip after luring the user to a malicious website.

Affected Products

  • Paperclip versions prior to 0.3.1
  • Deployments running the default local_trusted mode
  • Local developer environments exposing the Paperclip API on loopback

Discovery Timeline

  • 2026-08-21 - CVE-2026-77087 published to NVD
  • 2026-08-21 - Last updated in NVD database

Technical Details for CVE-2026-77087

Vulnerability Analysis

Paperclip exposes an HTTP API on the local host for developer workflows. In local_trusted mode, the service assumes any request reaching the loopback interface originates from the trusted user. The server does not inspect the Host header, so requests bound to arbitrary hostnames are accepted as long as they arrive on the listening port. This missing authorization check allows browser-mediated cross-origin requests to invoke privileged endpoints, including the process adapter that spawns operating system commands.

Root Cause

The root cause is a missing authorization control [CWE-862] combined with implicit trust of the network location. Paperclip treats loopback connectivity as proof of local, trusted origin. Because the Host header is not validated against an allow list, an attacker-controlled hostname that resolves to 127.0.0.1 passes all server-side checks. The process adapter endpoint compounds the impact by exposing command execution to any caller that satisfies the flawed trust model.

Attack Vector

The attack requires user interaction but no authentication. A developer visits an attacker-controlled webpage while Paperclip is running locally. The page loads a script from a domain whose DNS record has a very short time-to-live. After initial resolution, the attacker rebinds the domain to 127.0.0.1. Subsequent fetch or XMLHttpRequest calls from the still-open page reach Paperclip on loopback, but the browser considers them same-origin. The script then calls the process adapter API and executes arbitrary commands with the developer's privileges. See the GitHub Security Advisory and the VulnCheck Advisory on Paperclip for further technical detail.

Detection Methods for CVE-2026-77087

Indicators of Compromise

  • Unexpected child processes spawned by the Paperclip process on developer workstations
  • HTTP requests to the Paperclip listener carrying Host headers that do not match localhost or 127.0.0.1
  • Outbound DNS queries to newly registered domains with abnormally short TTL values preceding local process execution
  • Browser network logs showing cross-origin requests to the Paperclip API port

Detection Strategies

  • Inspect HTTP access logs on the Paperclip listener for Host headers referencing external domains rather than the loopback address
  • Correlate browser session activity with local Paperclip process adapter invocations to surface DNS rebinding chains
  • Alert on process trees where a developer tool such as Paperclip spawns shells, scripting interpreters, or package managers
  • Baseline expected API callers and flag unusual user agents or referrers hitting the local API

Monitoring Recommendations

  • Enable endpoint process telemetry on developer machines to record parent-child relationships involving Paperclip
  • Collect and centralize DNS resolver logs to identify short-TTL responses resolving to private address ranges
  • Monitor egress web traffic for pages that repeatedly query loopback services shortly after load

How to Mitigate CVE-2026-77087

Immediate Actions Required

  • Upgrade Paperclip to version 0.3.1 or later on every developer workstation and build agent
  • Restart any running Paperclip instances after upgrading to ensure the patched binary is loaded
  • Audit developer machines for signs of prior process adapter abuse while the vulnerability was exposed

Patch Information

Paperclip 0.3.1 fixes the issue by validating the Host header against an allow list of trusted local names. Refer to the GitHub Security Advisory GHSA-x8hx-rhr2-9rf7 for release notes and the fix commit.

Workarounds

  • Bind the Paperclip listener to a Unix domain socket or a non-routable interface that browsers cannot reach
  • Place a local reverse proxy in front of Paperclip that enforces strict Host header and Origin validation
  • Require an authentication token on every API request rather than relying on loopback trust
  • Disable Paperclip when it is not actively in use to shrink the exposure window
bash
# Configuration example: verify the installed Paperclip version and upgrade
paperclip --version
pip install --upgrade "paperclip>=0.3.1"
# Restrict the listener to loopback and require a token (example placeholder)
export PAPERCLIP_BIND=127.0.0.1
export PAPERCLIP_REQUIRE_TOKEN=1

Disclaimer: This content was generated using AI. While we strive for accuracy, please verify critical information with official sources.

Default Legacy - Prefooter | Experience the World’s Most Advanced Cybersecurity Platform

Experience the Most Advanced Cybersecurity Platform

See how the world’s most intelligent, autonomous cybersecurity platform can protect your organization today and into the future.