ThisBash - Claude Dispatch browser extension RCE (researcher disclosure)
Forever Security researcher Gal Weizman disclosed ThisBash vulnerability allowing browser extension on claude.ai to push prompts to Claude Desktop via Dispatch API, auto-clicking Allow button to execute…
What happened
Forever Security researcher Gal Weizman disclosed ThisBash vulnerability allowing browser extension on claude.ai to push prompts to Claude Desktop via Dispatch API, auto-clicking Allow button to execute commands without user interaction. Proof-of-concept demonstrated npm install -g add-colors-to-bash execution (opened Calculator app). Attack can reach other machines logged into same Claude account. Anthropic announced Dispatch retirement. No evidence of in-the-wild exploitation; marked borderline as researcher disclosure only.
Browser extension on claude.ai pushes prompts to Claude Desktop via Dispatch, auto-clicks Allow; enables RCE reaching other machines on same account
How it happened
Read the flow as text & sources
A catalog reconstruction of the supplied source account. Researcher disclosure; no evidence of in-the-wild exploitation; Anthropic retiring Dispatch feature
- 01 · Malicious extension installed
User installs malicious browser extension with claude.ai permissions
Hunt: A new or updated extension gains sensitive permissions and accesses conversation, cookie, or browsing data before unusual egress.
Extension installation does not prove theft. Endpoint application control is not a substitute for browser extension allowlisting.
Source for step 1 - 02 · Extension pushes prompt via Dispatch
Extension uses Dispatch API to push command prompt to Claude Desktop
Hunt: An agent ingests external content or an untrusted session identifier before a tool action or configuration change outside the approved task.
Requires application instrumentation and a custom field mapping. Input exposure does not prove that the model followed instructions, executed a tool or bypassed approval. Avoid collecting sensitive prompt bodies unless authorized.
Source for step 2 - 03 · Auto-click Allow button
Extension automatically clicks Allow to bypass user consent
Hunt: Untrusted content or a changed tool definition precedes a file read, command, or remote call outside the user's intended task.
Process logs cannot recover model intent. Remote connectors and in-process interpreters may produce no local child process. Prompt content alone is not a reliable signature.
Source for step 3 - 04 · Execute arbitrary command
Command executes on Claude Desktop (PoC: npm install -g add-colors-to-bash)
Source for step 4
Why it matters
Reported entry point: Malicious browser extension on claude.ai domain. Reported impact: Command executes on Claude Desktop (PoC: npm install -g add-colors-to-bash); reported targets include Claude Desktop users; theoretical risk only (no in-wild use); applicability to your setup has not been assessed, so verify affected versions, permissions, and configuration in the original report.
What you can do
Protection & prevention
Anthropic retiring Dispatch feature; users should update Claude Desktop when patch available; review browser extensions on claude.ai; implement extension permission review
Catalog defensive guidance based on the reported behavior. Confirm scope and current mitigations in the original source.
Detection queries
Draft query starters grouped by the attack stages where they apply.
Start with the logs you collect. Each hunt identifies its required visibility, field mapping and limits. Queries are catalog-authored drafts using reusable behavior profiles; test with known activity and your normal baseline before alerting.
01Malicious extension installedBrowser extension change / data accessCustom schema · Draft
User installs malicious browser extension with claude.ai permissions
What to look for
A new or updated extension gains sensitive permissions and accesses conversation, cookie, or browsing data before unusual egress.
Visibility: Defender-side logs where collected
Required logs
- Managed-browser extension inventory and update history
- Extension ID, permissions and provenance
- Browser / proxy / DLP evidence
Coverage limit. Extension installation does not prove theft. Endpoint application control is not a substitute for browser extension allowlisting.
Custom-schema candidates, not out-of-box detectors. Map your application / cloud / platform logs to EventType, Actor, Asset, Action, Target and CorrelationId. KQL uses a custom OpenAttacksAudit_CL table; SPL uses openattacks:audit. EventType values are proposed normalization names, not native vendor event names. Correlation and environment-specific alert logic remain to be built.
Sigma
title: Hunt - Browser extension change / data access
id: 0dd8a9d2-e0c8-4768-a49d-5fc2f09a6d7c
status: experimental
description: A new or updated extension gains sensitive permissions and accesses
conversation, cookie, or browsing data before unusual egress.
author: MagicSword OpenAttacks catalog
date: 2026-10-23
falsepositives:
- Authorized activity matching this broad hunt; baseline and investigate
before alerting.
level: low
logsource:
category: application
product: openattacks_audit
definition: Custom normalized audit schema; map EventType, Actor, Asset, Action,
Target and CorrelationId from the required source logs first.
detection:
selection:
EventType:
- browser.extension.install
- browser.extension.update
condition: selection
KQL
// SCHEMA TEMPLATE: create/map this custom table before use; not a built-in Sentinel table.
// OpenAttacksAudit_CL: TimeGenerated(datetime), EventType/Actor/Asset/Action/Target/CorrelationId(string).
// Selects candidates only. Apply the stage hypothesis and correlation steps before alerting.
OpenAttacksAudit_CL
| where TimeGenerated > ago(7d)
| where EventType in ("browser.extension.install", "browser.extension.update")
| project TimeGenerated, EventType, Actor, Asset, Action, Target, CorrelationId
| order by TimeGenerated descSPL
sourcetype="openattacks:audit"
EventType IN ("browser.extension.install", "browser.extension.update")
| table _time EventType Actor Asset Action Target CorrelationId
| sort - _timeInvestigation steps
- Diff extension versions and permissions, compare the extension ID against the allowlist, and inspect its update origin.
- Correlate access to conversations or credentials with destinations; enforce browser extension policy separately.
02Extension pushes prompt via DispatchUntrusted content enters an agent sessionCustom schema · Draft
Extension uses Dispatch API to push command prompt to Claude Desktop
What to look for
An agent ingests external content or an untrusted session identifier before a tool action or configuration change outside the approved task.
Visibility: Application / agent audit records
Required logs
- Input provenance, source type and request/session identifier
- Captured tool responses or retrieved content where collection is approved
- Subsequent tool calls, configuration changes and approval decisions
Coverage limit. Requires application instrumentation and a custom field mapping. Input exposure does not prove that the model followed instructions, executed a tool or bypassed approval. Avoid collecting sensitive prompt bodies unless authorized.
Custom-schema candidate for input exposure only. Map agent.input.ingest and agent.session.resume to your application request and session logs. CorrelationId must link subsequent actions. EventType values are proposed normalization names, not native vendor events.
Sigma
title: Hunt - Untrusted content enters an agent session
id: 2d89d7ae-9d44-4dc6-a95b-f9c8de7b8352
status: experimental
description: An agent ingests external content or an untrusted session
identifier before a tool action or configuration change outside the approved
task.
author: MagicSword OpenAttacks catalog
date: 2026-11-04
falsepositives:
- Authorized activity matching this broad hunt; baseline and investigate
before alerting.
level: low
logsource:
category: application
product: openattacks_audit
definition: Custom normalized audit schema; map EventType, Actor, Asset, Action,
Target and CorrelationId from the required source logs first.
detection:
selection:
EventType:
- agent.input.ingest
- agent.session.resume
condition: selection
KQL
// SCHEMA TEMPLATE: create/map this custom table before use; not a built-in Sentinel table.
// OpenAttacksAudit_CL: TimeGenerated(datetime), EventType/Actor/Asset/Action/Target/CorrelationId(string).
// Selects candidates only. Apply the stage hypothesis and correlation steps before alerting.
OpenAttacksAudit_CL
| where TimeGenerated > ago(7d)
| where EventType in ("agent.input.ingest", "agent.session.resume")
| project TimeGenerated, EventType, Actor, Asset, Action, Target, CorrelationId
| order by TimeGenerated descSPL
sourcetype="openattacks:audit"
EventType IN ("agent.input.ingest", "agent.session.resume")
| table _time EventType Actor Asset Action Target CorrelationId
| sort - _timeInvestigation steps
- Join the ingested input to later actions by session or request ID; preserve the original content as evidence.
- Compare the resulting action with the approved task and baseline. Receiving suspicious text alone is not successful injection.
03Auto-click Allow buttonUnexpected agent tool useCustom schema · Draft
Extension automatically clicks Allow to bypass user consent
What to look for
Untrusted content or a changed tool definition precedes a file read, command, or remote call outside the user's intended task.
Visibility: Defender-side logs where collected
Required logs
- Agent tool-call audit records with session / tool / destination
- Tool-definition and approval changes
- Endpoint process, file-access and network evidence where available
Coverage limit. Process logs cannot recover model intent. Remote connectors and in-process interpreters may produce no local child process. Prompt content alone is not a reliable signature.
Custom-schema candidates, not out-of-box detectors. Map your application / cloud / platform logs to EventType, Actor, Asset, Action, Target and CorrelationId. KQL uses a custom OpenAttacksAudit_CL table; SPL uses openattacks:audit. EventType values are proposed normalization names, not native vendor event names. Correlation and environment-specific alert logic remain to be built.
Sigma
title: Hunt - Unexpected agent tool use
id: c709e384-0e11-479e-ab4d-3a06e2431e4b
status: experimental
description: Untrusted content or a changed tool definition precedes a file
read, command, or remote call outside the user's intended task.
author: MagicSword OpenAttacks catalog
date: 2026-10-23
falsepositives:
- Authorized activity matching this broad hunt; baseline and investigate
before alerting.
level: low
logsource:
category: application
product: openattacks_audit
definition: Custom normalized audit schema; map EventType, Actor, Asset, Action,
Target and CorrelationId from the required source logs first.
detection:
selection:
EventType:
- agent.tool.call
- agent.tool.definition.change
condition: selection
KQL
// SCHEMA TEMPLATE: create/map this custom table before use; not a built-in Sentinel table.
// OpenAttacksAudit_CL: TimeGenerated(datetime), EventType/Actor/Asset/Action/Target/CorrelationId(string).
// Selects candidates only. Apply the stage hypothesis and correlation steps before alerting.
OpenAttacksAudit_CL
| where TimeGenerated > ago(7d)
| where EventType in ("agent.tool.call", "agent.tool.definition.change")
| project TimeGenerated, EventType, Actor, Asset, Action, Target, CorrelationId
| order by TimeGenerated descSPL
sourcetype="openattacks:audit"
EventType IN ("agent.tool.call", "agent.tool.definition.change")
| table _time EventType Actor Asset Action Target CorrelationId
| sort - _timeInvestigation steps
- Join input provenance, tool descriptions and calls by session ID and compare the action to the approved task.
- Inspect child processes and destinations; record whether execution was in-process, local, or through a remote SaaS connector.
Technical detailsTooling, impact & MITRE ATLAS
Attack context
- AI's role
- Agent / MCP as target
- Initial access
- Malicious browser extension on claude.ai domain
Tooling as reported
AI products, models and artifacts
- Claude Desktop
- Claude Dispatch API
Other tooling
- Browser extension
- npm
Targets and reported impact
- Targets / sectors
- Claude Desktop users; theoretical risk only (no in-wild use)
- Reported scale / exposure
- PoC only; Anthropic retiring Dispatch; no in-wild exploitation documented
Reported attribution
Researcher disclosure (Gal Weizman, Forever Security); no threat actor
Structured attribution
- Actor
- Researcher disclosure only
- Country
- Unknown
- Motivation
- unknown
- Attributed by
- Forever Security
- Attribution confidence
- low
Detection rules
- Claude Dispatch security monitoringGUIDANCE
Monitor browser extensions with claude.ai permissions; detect auto-clicking behaviors; log Dispatch API calls; update Claude Desktop when Dispatch retirement patch available; review extension permissions regularly
Activity timeline
- First seen
- 2026-11-06
- Last updated
- 2026-11-06
Updates
- 2026-11-06 — Forever Security
Gal Weizman discloses ThisBash vulnerability; Anthropic announces Dispatch retirement; no in-wild exploitation
Source defensive notes
Anthropic retiring Dispatch feature; users should update Claude Desktop when patch available; review browser extensions on claude.ai; implement extension permission review
MITRE ATLAS
AI-specific catalog annotations based on the reported behavior; these are not MITRE case-study assignments.
Browser extension could inject prompts to Claude Desktop via Dispatch Source
Auto-click Allow bypasses user consent for command execution Source
Sources & research credit
1 sourceResearch by Gal Weizman · Forever Security
Forever Security blog (Nov 6, 2026); researcher disclosure only; no threat actor or in-wild use; Anthropic announced Dispatch retirement
Record & downloads
- Source record name
- ThisBash - Claude Dispatch browser extension RCE (researcher disclosure)
- Entry ID
- thisbash-claude-dispatch-rce-2026-11
- Catalog review
- November 10, 2026
- Category
- attacks-on-ai
- AI involvement confidence
- confirmed