Skip to content
Submit incident
AII-1001-09441
ConfirmedA court, a regulator or the organization itself has stated the core facts, or the record cites a court or regulator source.

Path traversal in OpenClaw browser upload allows local file read

OpenClaw's browser upload feature contained a path traversal vulnerability that allowed authenticated attackers to read arbitrary files from the Gateway host by supplying absolute paths or traversal sequences. The vulnerability was patched in version 2026.2.14.

15 February 2026 Β· OpenClaw Β· Security and cyberattacks
15 February 2026
AII-1001-09441View source β†—
Quick facts
Organization
OpenClaw (subject)
Category
Security and cyberattacks
Country
United States
Source
github.com
Date
15 February 2026
Drafted by
Raidu Incident Detection

What happened

Authenticated attackers could read arbitrary files from the Gateway host by supplying absolute paths or path traversal sequences to the browser tool's upload action via POST /tools/invoke with {"tool":"browser","action":"upload",...} or POST /hooks/file-chooser.

The server passed upload paths to Playwright's setInputFiles() APIs without restricting them to a safe root, allowing the referenced files to be read from the local filesystem and contents exfiltrated by page JavaScript or via agent/browser snapshots.

Reported impact

The vulnerability enabled arbitrary local file read on the Gateway host, affecting confidentiality.

Index analysis

The vulnerability required authenticated access and the browser tool to be permitted by tool policy; in default setups the Gateway binds to loopback with a generated gateway token for onboarding.

Governance controls

Upload paths are now confined to OpenClaw's temp uploads root (DEFAULT_UPLOAD_DIR) and traversal/escape paths are rejected, as implemented in fix commit 3aa94af.

Classification

Organization
OpenClaw (subject)
Category
Security and cyberattacks
Industry
IT infrastructure and hosting
Country
United States
Harm status
Potential harm
Intent
Unintentional
Timing
Post-deployment
Harm types
Public interest
AI systems
OpenClaw

Sources and evidence

This record was researched and written by the Index, based on the following original reporting.

github.com
https://github.com/openclaw/openclaw/security/advisories/GHSA-cv7m-c9jx-vg7q
Aug 2026Claude AI agent discovers gym booking system vulnerability, cancels another user's reservationWell documentedSecurity and cyberattacksThe IndependentDiaspora Digital Media (DDM News) - Nigeria Breaking News, Africa and World News and Updates -The Indian Express+3Jun 2026Five malicious skills removed from OpenClaw ClawHub marketplaceConfirmedSecurity and cyberattacksUnit42Dark ReadingJun 2026OpenClaw AI agent vulnerabilities enable malicious skill packages to steal user dataUnder investigationSecurity and cyberattacksenglish.news.cnChina NewsApr 2026AI Agent Exploited Gym Booking API Authorization Flaw and Removed Member From WaitlistReportedSecurity and cyberattacksabc.net.auaffinda.comApr 2026Open-source AI agents expose security vulnerabilities in ChinaReportedSecurity and cyberattacksδΈœζ–Ήθ΄’ε―Œη½‘China News