Skip to main content

Summarize CVE-2026-45244

| EUVDEUVD-2026-30796 LOW
Missing Authorization (CWE-862)
2026-05-18 VulnCheck GHSA-67gq-6q8c-qqh6
2.1
CVSS 4.0 · Vendor: VulnCheck
View

Attack ChainAIDerived

Hypothetical attack flow derived from CVE metadata – not a confirmed exploit chain

  1. Access
    User enables extension automation feature
  2. Delivery
    Attacker crafts malicious page with injected prompt instructions
  3. Exploit
    User invokes Summarize on attacker-controlled content
  4. Execution
    Agent processes unescaped content delimiters as trusted instructions
  5. Persist
    Per-call approval step bypassed
  6. Impact
    Unauthorized browser automation actions execute

Vulnerability Assessment

Weakness
CWE
OWASP Top 10:2025
A01:2025 - Broken Access Control
ASVS 5.0
V8 - Authorization Authorization checks on every request
MITRE ATT&CK

1 technique across 1 tactic: Privilege Escalation

The technique list with ATT&CK links and the kill-chain view is part of the Basic plan and up.

Exploitation Conditions

The extension automation feature in Summarize must be explicitly enabled by the user - this is the single most important limiting condition, as the vulnerable code path is entirely gated by this non-default feature toggle. The user must interact with attacker-controlled content (UI:R per CVSS vector), meaning the user must actively trigger Summarize on a malicious page or summary. No attacker-side authentication or privileges are required (PR:N). Exploitation is scoped to browser-level actions within the extension's automation toolkit (navigation, debugger-backed actions) and cannot escalate beyond the browser sandbox (S:U). Versions at or above 0.15.2 contain the delimiter-escaping fix and are not exploitable via this vector.

Impact

CVSS scores this at 5.4 (Medium) with a vector of AV:N/AC:L/PR:N/UI:R/S:U/C:L/I:L/A:N. Network delivery with low complexity and no required attacker privileges increases theoretical reachability, but the mandatory user interaction (UI:R) and unchanged scope (S:U) meaningfully constrain real-world impact. The confidentiality and integrity impacts are both rated Low, consistent with browser-bounded automation actions rather than OS-level compromise. No EPSS data was provided in the input, and there is no CISA KEV listing - active exploitation is unconfirmed. The non-default requirement of the extension automation feature being explicitly enabled further limits exposure to a subset of the installed userbase. The CVSS vector is consistent with the vulnerability description; no conflicting signals are present. Overall, this is a real but bounded risk: moderate severity for users with automation enabled, low priority for users who have never activated that feature.

Exploit ScenarioDerived

A user with the Summarize extension installed and the extension automation feature enabled visits a web page crafted by an attacker to embed prompt-injection instructions within visible or hidden page content. When the user invokes Summarize on that page, the AI agent processes the malicious content, interprets the injected instructions as legitimate automation requests, and executes browser automation actions - such as navigating to a new URL or invoking debugger-accessible operations - without surfacing the per-call approval dialog that would otherwise require the user's explicit consent.

Monitor Track and reassess. Why: LOW severity (CVSS 0.1-3.9).
Patch available 0.0% exploitation probability (EPSS)

Get an e-mail when this CVE is exploited, patched or re-scored. Free account.

Alert me

AnalysisAIv1 – May 18, 2026

Missing authorization in the Summarize browser extension (versions prior to 0.15.1/0.15.2, CPE: cpe:2.3:a:steipete:summarize) allows remote unauthenticated attackers to execute browser automation actions - including navigation and debugger-backed operations - without triggering per-call user approval. Exploitation requires the extension automation feature to be enabled and the user to interact with attacker-controlled content (UI:R per CVSS), making this a prompt-injection-driven authorization bypass rather than a standalone remote attack. No public exploit has been identified at time of analysis, and the vendor released a patch in v0.15.2 as reported by VulnCheck.

RemediationAI

Upgrade to Summarize v0.15.2 or later, which includes the fix for untrusted prompt context and content delimiter escaping introduced in commit e64fe3ecd1bb4fdc181dcfa88c96b9e1914ced0e via PR https://github.com/steipete/summarize/pull/219. The patched npm packages @steipete/summarize@0.15.2 and @steipete/summarize-core@0.15.2 are available at their respective npm registry URLs as of 2026-05-17. The release is available at https://github.com/steipete/summarize/releases/tag/v0.15.2 for both Chrome and Firefox extension distributions. As a compensating control prior to upgrading, explicitly disabling the extension automation feature eliminates the attack surface entirely, since this feature must be enabled for exploitation - the trade-off is loss of all browser automation functionality including navigation and debugger-backed actions. No other workarounds are identified from available data.

Affected VersionsENISA

summarize 0 <0.15.1
summarize patch: e64fe3ecd1bb4fdc181dcfa88c96b9e1914ced0e

Affected Packages

npm @steipete/summarize fix: 0.15.0
euvd summarize fix: 0.15.1

Source: OSV (Open Source Vulnerabilities)

Affects your stack

Affects 2 packages in npm, euvd

Checking this CVE against your own stacks and servers is part of the Basic plan and up.

Technical ContextAI · show details

Summarize is an AI-powered browser extension and CLI tool (CPE: cpe:2.3:a:steipete:summarize) that processes web page and document content through an AI agent and exposes optional extension automation capabilities including browser navigation and Chrome DevTools Protocol (debugger-backed) actions. The root cause is CWE-862 (Missing Authorization): when the extension automation feature is enabled, the system feeds page or summary content into the agent without escaping content delimiters, creating a prompt injection pathway. Attacker-controlled content can embed instructions that cause the agent to invoke automation tools, bypassing the per-call approval gate that is supposed to require explicit user consent before each automation action executes. The v0.15.2 fix explicitly addresses this by escaping untrusted prompt context and content delimiters (per release notes: 'Core prompts: escape untrusted prompt context/content delimiters'), confirming that insufficient input sanitization upstream of the authorization check was the mechanism of bypass.

Know when Summarize gets its next vulnerability

  • Watch products and get an email when something new is published
  • Check this CVE against your own stack and servers
  • Exploit and PoC references, threat intelligence from 22 sources

No credit card · 7-day full trial

Severity by source

Vendor (VulnCheck) PRIMARY 2.1 LOW
CVSS:4.0/AV:N/AC:L/AT:P/PR:N/UI:A/VC:N/VI:N/VA:N/SC:L/SI:L/SA:N

Primary rating from Vendor (VulnCheck) · only source for this CVE.

CVSS VectorVendor: VulnCheck

Attack Vector
Network
Attack Complexity
Low
Attack Requirements
Present
Privileges Required
None
User Interaction
Active
Vulnerable system: Confidentiality
None
Vulnerable system: Integrity
None
Vulnerable system: Availability
None
Subsequent system: Confidentiality
Low
Subsequent system: Integrity
Low
Subsequent system: Availability
None

Lifecycle Timeline

2
Severity Changed
May 18, 2026 - 20:22 NVD
MEDIUM LOW
CVSS changed
May 18, 2026 - 20:22 NVD
5.4 (MEDIUM) 2.1 (LOW)

DescriptionCVE.org

Summarize prior to 0.15.1 contains a missing authorization vulnerability that allows attackers to execute browser automation actions without per-call user approval when the extension automation feature is enabled. Attackers can influence the agent through malicious page or summary content to invoke enabled extension automation tools such as navigation or debugger-backed actions, bypassing the final user approval step when a user interacts with attacker-controlled content.

Threat Intelligence Overview

EPSS Score
0.0%
Top 93% exploitable
Data Quality
7 sources
Confidence: High

How this decision was reached

The engine reads each signal below and applies a fixed list of rules from top to bottom. The first rule that matches sets the recommended action – like a statement, every line is shown.

SignalValue
Severity LOW
CISA KEV No
Public POC No
EPSS 0.0%
Patch available Yes
Active threat No
Deciding rule
LOW severity (CVSS 0.1-3.9)
What every action level means
Emergency Immediate action required. Criteria: CRITICAL + in CISA KEV (actively exploited) without patch, OR CRITICAL + EPSS > 30% without patch, OR CRITICAL + active threat signals without patch.
Act Now Prioritize within 24-48h. Criteria: in KEV with patch available, OR CRITICAL + public POC, OR HIGH + EPSS > 10%, OR CRITICAL + active threat, OR any CRITICAL severity.
This Week Schedule remediation this week. Criteria: HIGH severity (CVSS 7.0-8.9). Patch within normal change window.
This Month Plan within 30 days. Criteria: MEDIUM severity (CVSS 4.0-6.9), lower exploitation probability. Include in regular patch cycle.
Monitor Track and reassess. Criteria: LOW severity (CVSS 0.1-3.9), minimal exploitation risk. No immediate action needed.

SSVC AssessmentENISA

SSVC Decision· derived
Defer
Exploitation
None
System Exposure· derived
Open
Automatable
No
Human Impact· derived
Low
Technical Impact
Partial

Priority Score

10
Low Medium High Critical
KEV: 0
EPSS: 0
CVSS: +10
PoC: 0
Weaponized: 0
Threat intel: 0
Gov advisory: 0

Attack Technique

Authentication bypass attacks exploit flaws in the verification mechanisms that control access to systems and applications.

This site uses cookies essential for authentication and security. No tracking or analytics cookies are used. Privacy Policy