A Leader in the 2026 Gartner® Magic Quadrant™ for Endpoint Protection. Six years running.Six years. Gartner® Magic Quadrant™ Leader.Find Out Why
Experiencing a Breach?Blog
Get StartedContact Us
SentinelOne
  • Platform
    Platform Overview
    • Singularity Platform
      Welcome to Integrated Enterprise Security
    • AI for Security
      Leading the Way in AI-Powered Security Solutions
    • Securing AI
      Accelerate AI Adoption with Secure AI Tools, Apps, and Agents.
    • How It Works
      The Singularity XDR Difference
    • Singularity Marketplace
      One-Click Integrations to Unlock the Power of XDR
    • Pricing & Packaging
      Comparisons and Guidance at a Glance
    Data & AI
    • Purple AI
      Accelerate SecOps with Generative AI
    • Singularity Hyperautomation
      Easily Automate Security Processes
    • AI-SIEM
      The AI SIEM for the Autonomous SOC
    • AI Data Pipelines
      Security Data Pipeline for AI SIEM and Data Optimization
    • Singularity Data Lake
      AI-Powered, Unified Data Lake
    • Singularity Data Lake for Log Analytics
      Seamlessly Ingest Data from On-Prem, Cloud or Hybrid Environments
    Endpoint Security
    • Singularity Endpoint
      Autonomous Prevention, Detection, and Response
    • Singularity XDR
      Native & Open Protection, Detection, and Response
    • Singularity RemoteOps Forensics
      Orchestrate Forensics at Scale
    • Singularity Threat Intelligence
      Comprehensive Adversary Intelligence
    • Singularity Vulnerability Management
      Application & OS Vulnerability Management
    • Singularity Identity
      Identity Threat Detection and Response
    Cloud Security
    • Singularity Cloud Security
      Block Attacks with an AI-Powered CNAPP
    • Singularity Cloud Native Security
      Secure Cloud and Development Resources
    • Singularity Cloud Workload Security
      Real-Time Cloud Workload Protection Platform
    • Singularity Cloud Data Security
      AI-Powered Threat Detection for Cloud Storage
    • Singularity Cloud Security Posture Management
      Detect and Remediate Cloud Misconfigurations
    Securing AI
    • Prompt Security
      Secure AI Tools Across Your Enterprise
  • Why SentinelOne?
    Why SentinelOne?
    • Why SentinelOne?
      Cybersecurity Built for What’s Next
    • Our Customers
      Trusted by the World’s Leading Enterprises
    • Industry Recognition
      Tested and Proven by the Experts
    • About Us
      The Industry Leader in Autonomous Cybersecurity
    Compare SentinelOne
    • Arctic Wolf
    • Broadcom
    • CrowdStrike
    • Cybereason
    • Microsoft
    • Palo Alto Networks
    • Sophos
    • Splunk
    • Trellix
    • Trend Micro
    • Wiz
    Verticals
    • Energy
    • Federal Government
    • Finance
    • Healthcare
    • Higher Education
    • K-12 Education
    • Manufacturing
    • Retail
    • State and Local Government
  • Services
    Managed Services
    • Managed Services Overview
      Wayfinder Threat Detection & Response
    • Threat Hunting
      World-Class Expertise and Threat Intelligence
    • Managed Detection & Response
      24/7/365 Expert MDR Across Your Entire Environment
    • Incident Readiness & Response
      DFIR, Breach Readiness, & Compromise Assessments
    Support, Deployment, & Health
    • Technical Account Management
      Customer Success with Personalized Service
    • SentinelOne GO
      Guided Onboarding & Deployment Advisory
    • SentinelOne University
      Live and On-Demand Training
    • Services Overview
      Comprehensive Solutions for Seamless Security Operations
    • SentinelOne Community
      Community Login
  • Partners
    Our Network
    • MSSP Partners
      Succeed Faster with SentinelOne
    • Singularity Marketplace
      Extend the Power of S1 Technology
    • Cyber Risk Partners
      Enlist Pro Response and Advisory Teams
    • Technology Alliances
      Integrated, Enterprise-Scale Solutions
    • SentinelOne for AWS
      Hosted in AWS Regions Around the World
    • Channel Partners
      Deliver the Right Solutions, Together
    • SentinelOne for Google Cloud
      Unified, Autonomous Security Giving Defenders the Advantage at Global Scale
    • Partner Locator
      Your Go-to Source for Our Top Partners in Your Region
    Partner Portal→
  • Resources
    Resource Center
    • Case Studies
    • Data Sheets
    • eBooks
    • Reports
    • Videos
    • Webinars
    • Whitepapers
    • Events
    View All Resources→
    Blog
    • Feature Spotlight
    • For CISO/CIO
    • From the Front Lines
    • Identity
    • Cloud
    • macOS
    • SentinelOne Blog
    Blog→
    Tech Resources
    • SentinelLABS
    • Ransomware Anthology
    • Cybersecurity 101
  • About
    About SentinelOne
    • About SentinelOne
      The Industry Leader in Cybersecurity
    • Investor Relations
      Financial Information & Events
    • SentinelLABS
      Threat Research for the Modern Threat Hunter
    • Careers
      The Latest Job Opportunities
    • Press & News
      Company Announcements
    • Cybersecurity Blog
      The Latest Cybersecurity Threats, News, & More
    • FAQ
      Get Answers to Our Most Frequently Asked Questions
    • DataSet
      The Live Data Platform
    • S Foundation
      Securing a Safer Future for All
    • S Ventures
      Investing in the Next Generation of Security, Data and AI
  • Pricing
Get StartedContact Us
CVE Vulnerability Database
Vulnerability Database/CVE-2025-53097

CVE-2025-53097: Roo Code Information Disclosure Flaw

CVE-2025-53097 is an information disclosure vulnerability in Roocode Roo Code that allows attackers to read sensitive files outside the workspace. This article covers technical details, affected versions, impact, and mitigation.

Published: May 4, 2026

CVE-2025-53097 Overview

CVE-2025-53097 affects Roo Code, an AI-powered autonomous coding agent extension for Visual Studio Code. Versions prior to 3.20.3 contain a flaw in the search_files tool that ignores the user setting restricting reads to the VS Code workspace. An attacker capable of injecting a prompt into the agent can read sensitive files outside the workspace and exfiltrate the contents by writing them into a JSON schema. When schema fetching is enabled (the default), the write triggers an outbound network request without prompting the user. The maintainers fixed the issue in version 3.20.3.

Critical Impact

Prompt injection attacks against Roo Code can read arbitrary files outside the VS Code workspace and exfiltrate them via automatic JSON schema network fetches.

Affected Products

  • Roo Code VS Code extension versions prior to 3.20.3
  • Deployments with default JSON schema fetching enabled in VS Code
  • Roocode Roo Code (cpe:2.3:a:roocode:roo_code)

Discovery Timeline

  • 2025-06-27 - CVE-2025-53097 published to NVD
  • 2025-09-15 - Last updated in NVD database

Technical Details for CVE-2025-53097

Vulnerability Analysis

The vulnerability is classified as Improper Neutralization of Special Elements in Output Used by a Downstream Component [CWE-74]. Roo Code exposes a search_files tool to its agent loop. Users can configure the agent to limit reads to the active VS Code workspace. Prior to 3.20.3, search_files did not enforce this boundary check. An attacker who controls agent prompts, through indirect prompt injection in source files, documentation, or external content, can direct the agent to enumerate paths outside the workspace.

The exfiltration channel chains the read primitive with VS Code's automatic JSON schema resolution. Writing a remote $schema URL into a JSON file causes VS Code to issue an HTTP request to fetch the schema. By embedding stolen file contents into the schema URL or related fields, the attacker triggers an outbound request that leaks data without user approval.

Root Cause

The searchFilesTool implementation in src/core/tools/searchFilesTool.ts did not call isPathOutsideWorkspace to gate access against the alwaysAllowReadOnlyOutsideWorkspace setting. The auto-approve logic that protects other read tools was missing from this code path.

Attack Vector

Exploitation requires the attacker to deliver a prompt to the agent. Prompt injection vectors include malicious comments in third-party repositories, README files, dependency documentation, or any content the agent ingests. Once the agent executes, it reads sensitive files such as ~/.ssh/id_rsa or cloud credentials and writes them into a JSON schema reference, triggering automatic outbound retrieval by VS Code.

typescript
// Patch excerpt: src/core/tools/searchFilesTool.ts
 import { getReadablePath } from "../../utils/path"
 import { isPathOutsideWorkspace } from "../../utils/pathUtils"
 import { regexSearchFiles } from "../../services/ripgrep"
-import { t } from "../../i18n"

 export async function searchFilesTool(
 	cline: Task,

Source: Roo-Code commit 10b2fb3

A companion patch introduces an alwaysAllowWriteProtected setting and an isProtected message flag to block auto-approving edits to configuration files:

typescript
// Patch excerpt: packages/types/src/global-settings.ts
 	alwaysAllowReadOnlyOutsideWorkspace: z.boolean().optional(),
 	alwaysAllowWrite: z.boolean().optional(),
 	alwaysAllowWriteOutsideWorkspace: z.boolean().optional(),
+	alwaysAllowWriteProtected: z.boolean().optional(),
 	writeDelayMs: z.number().optional(),

Source: Roo-Code commit 7d0b22f

Detection Methods for CVE-2025-53097

Indicators of Compromise

  • Outbound HTTP/HTTPS requests from VS Code processes to attacker-controlled domains referenced in $schema fields.
  • Roo Code agent transcripts showing search_files invocations with absolute paths outside the configured workspace.
  • Newly created or modified .json files containing unfamiliar $schema URLs.

Detection Strategies

  • Audit Roo Code task logs for search_files calls referencing paths matching ~, /etc, /home, or user credential directories.
  • Inspect installed Roo Code extension version and flag any host running a release earlier than 3.20.3.
  • Monitor developer endpoints for VS Code child processes initiating connections to non-allowlisted schema hosts.

Monitoring Recommendations

  • Forward VS Code and extension telemetry to a centralized log store and alert on anomalous schema-fetch destinations.
  • Track changes to JSON files in repositories where Roo Code is active, focusing on $schema URL modifications.
  • Review agent prompt sources, including dependency README files and issue trackers, for embedded instructions targeting search_files.

How to Mitigate CVE-2025-53097

Immediate Actions Required

  • Upgrade Roo Code to version 3.20.3 or later across all developer workstations.
  • Disable JSON schema downloads in VS Code by setting json.schemaDownload.enable to false until updates are verified.
  • Restrict Roo Code to operate only on trusted repositories and review auto-approve settings for read and write tools.

Patch Information

The fix is delivered in Roo Code 3.20.3. Commit 10b2fb3 adds the workspace boundary check to searchFilesTool, and commit 7d0b22f introduces protections against auto-approving edits to configuration files. See the GitHub Security Advisory GHSA-wr2q-46pg-f228 for full details.

Workarounds

  • Set alwaysAllowReadOnlyOutsideWorkspace to false and require manual approval for all read operations.
  • Disable VS Code automatic JSON schema fetching to remove the exfiltration channel.
  • Run the IDE under an account without access to sensitive files such as SSH keys or cloud credentials.
bash
# VS Code settings.json - reduce exposure until patched
{
  "json.schemaDownload.enable": false,
  "roo-code.alwaysAllowReadOnlyOutsideWorkspace": false,
  "roo-code.alwaysAllowWriteOutsideWorkspace": false
}

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

  • Vulnerability Details
  • TypeInformation Disclosure

  • Vendor/TechRoocode

  • SeverityHIGH

  • CVSS Score7.5

  • EPSS Probability0.32%

  • Known ExploitedNo
  • CVSS Vector
  • CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:N/A:N
  • Impact Assessment
  • ConfidentialityLow
  • IntegrityNone
  • AvailabilityNone
  • CWE References
  • CWE-74
  • Vendor Resources
  • GitHub Commit Update

  • GitHub Commit Update

  • GitHub Security Advisory
  • Related CVEs
  • CVE-2025-65946: Roocode Roo Code RCE Vulnerability

  • CVE-2025-58371: Roocode Roo Code RCE Vulnerability

  • CVE-2025-58372: Roocode Roo Code RCE Vulnerability

  • CVE-2025-53536: Roocode Roo Code RCE Vulnerability
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.

Try SentinelOne
  • Get Started
  • Get a Demo
  • Product Tour
  • Why SentinelOne
  • Pricing & Packaging
  • FAQ
  • Contact
  • Contact Us
  • Customer Support
  • SentinelOne Status
  • Language
  • Platform
  • Singularity Platform
  • Singularity Endpoint
  • Singularity Cloud
  • Singularity AI-SIEM
  • Singularity Identity
  • Singularity Marketplace
  • Purple AI
  • Services
  • Wayfinder TDR
  • SentinelOne GO
  • Technical Account Management
  • Support Services
  • Verticals
  • Energy
  • Federal Government
  • Finance
  • Healthcare
  • Higher Education
  • K-12 Education
  • Manufacturing
  • Retail
  • State and Local Government
  • Cybersecurity for SMB
  • Resources
  • Blog
  • Labs
  • Case Studies
  • Videos
  • Product Tours
  • Events
  • Cybersecurity 101
  • eBooks
  • Webinars
  • Whitepapers
  • Press
  • News
  • Ransomware Anthology
  • Company
  • About Us
  • Our Customers
  • Careers
  • Partners
  • Legal & Compliance
  • Security & Compliance
  • Investor Relations
  • S Foundation
  • S Ventures

©2026 SentinelOne, All Rights Reserved.

Privacy Notice Terms of Use

English