# Unpatched Claude for Chrome Extension Flaws Enable Unauthorized Account Actions via Fake Clicks and Permission Bypass

> Manifold Security researchers reported two unpatched vulnerabilities in Anthropic's Claude for Chrome browser extension (through v1.0.80): a missing event.isTrusted validation in the content script's click handler that lets another extension forge a click and silently trigger hardcoded agentic prompts (Gmail, Google Docs, Calendar), and a ?skipPermissions=true URL parameter that grants the side panel a privileged, no-consent mode. Both remain reproducible eight releases after May 2026 disclosure, as of July 2026, and follow the earlier related 'ClaudeBleed' extension-trust flaw.

- **Published:** 2026-07-14T00:00:00Z
- **Last reviewed:** 2026-07-14T00:00:00Z
- **Canonical:** https://intel.threadlinqs.com/threat/TL-2026-1318
- **Card image:** https://intel.threadlinqs.com/og/threat/threat-intel-claude-for-chrome-istrusted-flaws-tl-2026-1318.webp?v=b0a1f308 (Threadlinqs threat intel: Claude for Chrome skipPermissions Flaws: Detection & IOCs (TL-2026-1318) — Critical severity, 19 IOCs, 9 detection rules.)
- **ID:** TL-2026-1318
- **Severity:** CRITICAL (CVSS 9.6)
- **Category:** VULNERABILITY
- **Status:** ACTIVE
- **Detections:** 9 (1 shown free below) · **IOCs:** 19

## Description

Anthropic's Claude for Chrome extension bridges the claude.ai LLM to a user's browser session via a privileged messaging interface (`externally_connectable` in the extension manifest) and a content script that listens for UI events to drive an agentic workflow. Manifold Security (led by Ax Sharma) identified and reported two distinct, unpatched weaknesses to Anthropic on May 21, 2026.

The first flaw is a click-validation bypass: the content script's onboarding/action click handler forwards matching UI events to Claude's side panel without checking the browser-native `event.isTrusted` property, which distinguishes genuine user-generated input events from script-dispatched synthetic ones. Because any other Chrome extension with script access to the claude.ai origin can dispatch a synthetic click event, an attacker-controlled extension can execute roughly six lines of JavaScript to simulate a click and silently trigger one of nine hardcoded prompts -- for example, reading and interacting with Gmail (including clicking unsubscribe links), opening and reading comments on the user's most recent Google Docs, or scanning and modifying Google Calendar entries. In Claude's default 'Ask before acting' mode a confirmation popup still appears, but when a user has enabled the extension's agentic 'Act without asking' mode, the forged click drives the action with no visible approval step -- the scenario the source article assigns a CVSS 9.6 (Critical) rating.

The second flaw is a URL-parameter-based privilege escalation: loading the extension's side panel with a `?skipPermissions=true` query parameter drops it into a privileged, no-consent execution mode without requiring any user gesture. Anthropic disputed the severity of this finding on the basis that presently only the extension itself constructs URLs containing this parameter, so no external actor can reach it directly. Manifold characterized it as a 'ticking time bomb': any future unrelated bug in the extension -- an XSS flaw, a new message handler, or a regression -- could let external code construct a `skipPermissions=true` URL and instantly obtain silent, full-account access.

Both issues were reported May 21, 2026. Anthropic acknowledged the reports within a day but disputed their severity, arguing the synthetic-click issue was already covered by existing internal documentation about the extension's threat model and that the URL parameter posed no external risk. Anthropic's internal tracking issue was marked 'Resolved' by June 9, 2026, without either fix shipping. Version 1.0.80, released July 7, 2026, still contained both flaws when Manifold re-verified, meaning the vulnerabilities persisted across eight point releases from initial disclosure.

These flaws follow a related, earlier disclosure by LayerX Security researcher Aviad Gispan, publicly documented as 'ClaudeBleed' (reported to Anthropic April 27, 2026; public disclosure May 8, 2026). ClaudeBleed exploited the same `externally_connectable` trust design flaw -- the extension validated the claude.ai *origin* of a message but not the *execution context* that actually sent it, letting any other installed extension (even one without special permissions) impersonate legitimate claude.ai script traffic and issue commands to the Claude agent. Anthropic's May 6, 2026 'partial' fix (v1.0.70) introduced an approval flow for privileged actions but did not address the root cause; Gispan demonstrated the fix could be bypassed by switching the agent into its 'privileged' mode without notifying the user. Researchers observed two bypass techniques recurring across both the ClaudeBleed and Manifold findings: 'Approval Looping' (programmatically forging repeated 'Yes, proceed' responses to satisfy confirmation prompts) and 'Perception Manipulation' (dynamically altering visible DOM/UI elements so the agent misreads a malicious action as benign). Manifold's Ax Sharma noted the core lesson is that 'monitoring AI agents at the prompt layer is fundamentally insufficient' when the agent's perceived environment itself can be manipulated to make malicious actions appear legitimate.

No CVE has been assigned to either the isTrusted click-validation bug or the skipPermissions privilege-escalation bug as of the July 14, 2026 reporting date. The flaws map to OWASP Top 10 for LLM Applications categories LLM01 (Prompt Injection) and LLM06 (Excessive Agency), reflecting the broader pattern of confused-deputy weaknesses in agentic browser extensions where an AI agent's privileged action surface can be reached and steered without genuine user authorization.

## MITRE ATT&CK

- T1199 Trusted Relationship
- T1059.007 JavaScript
- T1176 Software Extensions
- T1548 Abuse Elevation Control Mechanism
- T1685 Disable or Modify Tools
- T1036 Masquerading
- T1114.001 Local Email Collection
- T1213 Data from Information Repositories
- T1119 Automated Collection
- T1567.002 Exfiltration to Cloud Storage
- T1565 Data Manipulation
- T1185 Browser Session Hijacking
- T1213.003 Code Repositories
- T1534 Internal Spearphishing
- T1071.001 Web Protocols

## Free detection (Splunk SPL, Microsoft KQL, Sigma)

**Chrome Extension Synthetic Click Forgery Against Claude for Chrome (isTrusted=false)** (`TL-2026-1318-SPL-001`)

Detects a browser extension activity/audit log event where a click forwarded to the Claude for Chrome content-script handler is flagged isTrusted=false, indicating a sibling extension forged the event to silently trigger a hardcoded agentic prompt (Gmail/Docs/Calendar).

- **Languages:** Splunk SPL, Microsoft KQL, Sigma
- **ATT&CK:** T1199, T1059.007
- **Data sources:** browser_extension_activity_log, endpoint_edr_browser_telemetry
- **Severity:** critical
- **Confidence:** high

### Splunk SPL

```spl
index=browser_security sourcetype="chrome:extension:activity" EventType="synthetic_click" isTrusted=false ForwardedTo="*claude-for-chrome*"
| eval risk="exploitation-flaw1-isTrusted-bypass"
| table _time, host, user, source_extension, target_extension, forwarded_prompt, isTrusted, risk
| sort -_time
```

### Microsoft KQL

```kql
let lookback = 24h;
BrowserExtensionEvents
| where TimeGenerated > ago(lookback)
| where EventType == "synthetic_click"
| where isTrusted == false
| where ForwardedTo has "claude-for-chrome"
| project TimeGenerated, DeviceName, AccountName, SourceExtensionId, TargetExtension, ForwardedPrompt, isTrusted
```

### Sigma

```yaml
title: Chrome Extension Synthetic Click / isTrusted Bypass Against Claude for Chrome
id: 3f9a1e0c-1b2d-4a7e-9c3f-1a2b3c4d5e01
status: experimental
description: Detects a script-dispatched (non-trusted) click event forwarded to Claude for Chrome's content-script handler, indicating exploitation of the missing event.isTrusted validation (CWE-346).
references:
    - https://cybersecuritynews.com/claude-for-chrome-vulnerability/
    - https://www.securityweek.com/unpatched-claude-for-chrome-flaw-lets-extensions-read-gmail-calendar/
author: Threadlinqs Team
date: 2026-07-14
tags:
    - attack.initial-access
    - attack.t1199
    - attack.execution
    - attack.t1059.007
    - cwe.346
logsource:
    category: application
    product: browser
detection:
    selection:
        EventType: 'synthetic_click'
        isTrusted: false
        ForwardedTo|contains: 'claude-for-chrome'
    condition: selection
falsepositives:
    - Legitimate automated UI testing frameworks driving extension interfaces
level: high
```

**Known false positives:**

- Legitimate automated extension-testing frameworks (e.g. Selenium/Puppeteer extension harnesses) that dispatch synthetic click events during CI

**Tuning notes:** If the org whitelists specific QA/testing extension IDs, exclude those source_extension values; correlate with forwarded_prompt values matching the 9 hardcoded agentic prompts (Gmail/Docs/Calendar) to reduce noise on generic click telemetry.

### 8 more detection rules (Splunk SPL, Microsoft KQL, Sigma), unlock with TL Creds

- externally_connectable Message Origin/Execution-Context Mismatch (ClaudeBleed Confused-Deputy Pattern) (locked: unlock with TL Creds)
- Perception Manipulation — Sensitive-Field DOM Label Stripping Prior to Privileged Agent Action (locked: unlock with TL Creds)
- Agentic Extension Gmail/Drive/Calendar/GitHub Action Without Corresponding User-Confirmation Event (locked: unlock with TL Creds)
- Sigma: Claude for Chrome Synthetic Click / isTrusted Bypass (locked: unlock with TL Creds)
- Sigma: Approval-Looping Confirmation Forgery (locked: unlock with TL Creds)
- Sigma: skipPermissions URL-Parameter Privileged Mode Entry (locked: unlock with TL Creds)
- Approval-Looping Confirmation Forgery Against Claude for Chrome Privileged Action Prompt (locked: unlock with TL Creds)
- skipPermissions URL-Parameter Privileged Mode Entry on Claude for Chrome Side Panel (locked: unlock with TL Creds)

Sign up free to unlock the detection rules: https://threadlinqs.com/?modal=signup

## Sources

- [Claude for Chrome Vulnerability](https://cybersecuritynews.com/claude-for-chrome-vulnerability/)
- [Unpatched Claude for Chrome Flaw Lets Extensions Read Gmail, Calendar](https://www.securityweek.com/unpatched-claude-for-chrome-flaw-lets-extensions-read-gmail-calendar/)
- [Vulnerability in Claude Extension for Chrome Exposes AI Agent to Takeover](https://www.securityweek.com/vulnerability-in-claude-extension-for-chrome-exposes-ai-agent-to-takeover/)
- [Flaw in Claude's Chrome extension allowed 'any' other plugin to hijack victims' AI](https://cyberscoop.com/claude-chrome-extension-allows-plugins-to-hijack-ai/)
- [Claude in Chrome is taking orders from the wrong extensions](https://www.csoonline.com/article/4168867/claude-in-chrome-is-taking-orders-from-the-wrong-extensions.html)
- [Claude's Chrome Extension Vulnerability Allows Malicious Extensions to Steal Gmail and Drive Data](https://cybersecuritynews.com/claudes-chrome-extension-vulnerability/)

## Full data

The other 8 detection rules (Splunk SPL / Microsoft KQL / Sigma) unlock with TL Creds; a free account includes daily creds: https://threadlinqs.com/?modal=signup
Indicator values unlock with TL Creds, at a lower price on Red and Purple; AI agents can query rules and IOCs through the Threadlinqs MCP server (Purple): https://intel.threadlinqs.com/mcp

Canonical: https://intel.threadlinqs.com/threat/TL-2026-1318
