opencode remote MCP OAuth command execution
A malicious remote MCP server can return an OAuth URL that runs commands via PowerShell when opencode opens it on Windows or WSL. No prompt injection is needed.
What happened
Malicious or compromised remote MCP server returns 401 and OAuth discovery with attacker-controlled authorization_endpoint. opencode passes that URL to npm open; on Windows/WSL open 10.x builds a PowerShell -EncodedCommand without escaping the target, so $(...) in an https URL runs before the browser launches. No prompt injection required.
How it happened
Read the flow as text & sources
A catalog reconstruction of the supplied source account. Unknown steps remain explicit; an edge describes the reported relationship between steps.
- 01 · Developer connects a remote MCP server
The server is attacker-controlled or later compromised.
Hunt: A repository, dependency or agent changes MCP commands, automatic-task settings, tool approval or model-routing configuration before new execution.
A file-write event alone cannot establish what changed. Legitimate setup also changes these files; obtain a diff and authorization context.
Source for step 1 - 02 · OAuth discovery supplies an attacker URL
A 401 response leads to an attacker-controlled authorization endpoint.
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 2 - 03 · URL enters PowerShell through open()
Affected Windows or WSL browser-opening code fails to escape the target.
Hunt: A script host or shell runs encoded content, a remote URL, or inline script syntax, then downloads or launches a payload.
Encoded commands and remote administration can be legitimate. This does not cover every script host, in-process execution, or non-Windows branch.
MagicSword policy opportunity · The reported Windows / WSL PowerShell helper, only where execution policy restricts it. Policy is not a replacement for fixing unsafe URL handling.
Source for step 3 - 04 · Commands execute before the browser opens
The injection occurs in the URL-opening path without requiring prompt injection.
Hunt: A script host or shell runs encoded content, a remote URL, or inline script syntax, then downloads or launches a payload.
Encoded commands and remote administration can be legitimate. This does not cover every script host, in-process execution, or non-Windows branch.
Source for step 4
Why it matters
Reported entry point: Developer connects coding agent to attacker-controlled or later-compromised remote MCP server. Reported impact: The injection occurs in the URL-opening path without requiring prompt injection; reported targets include Windows and WSL developers using opencode (~3M weekly npm downloads cited by Pillar).
Check these source-linked conditions against your setup.
opencode opens an untrusted MCP OAuth authorization URL through the reported Windows execution path.
Conditions to check- Confirm that the installed version is affected by the historical report; selecting a product does not establish its version or patch status.
- The report describes Windows or WSL and a malicious or intercepted remote MCP OAuth server.
Source-linked applicability guidance. Confirm current scope in the original report.
What you can do
Protection & prevention
Check the fixed version and review remote MCP endpoints before opening their OAuth authorization URLs.
Upgrade opencode; inventory installations (no CVE for scanners); pull unvetted remote MCP; argv not shell for URLs; MagicSword prevent unapproved second-stage binaries
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.
01Developer connects a remote MCP serverAgent trust / configuration changesCustom schema · Draft
The server is attacker-controlled or later compromised.
What to look for
A repository, dependency or agent changes MCP commands, automatic-task settings, tool approval or model-routing configuration before new execution.
Compare the remote server origin and connection settings with the approved inventory before examining the OAuth execution stage.
Visibility: Defender-side logs where collected
Required logs
- File changes and before/after content for MCP / IDE configuration
- Repository audit / pull-request history
- Agent startup and tool-invocation logs
Coverage limit. A file-write event alone cannot establish what changed. Legitimate setup also changes these files; obtain a diff and authorization context.
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 - Agent trust / configuration changes
id: 2dfc8034-d66b-42a7-a699-a76b2ef872ac
status: experimental
description: A repository, dependency or agent changes MCP commands,
automatic-task settings, tool approval or model-routing configuration before
new execution.
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.config.write
- agent.approval.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.config.write", "agent.approval.change")
| project TimeGenerated, EventType, Actor, Asset, Action, Target, CorrelationId
| order by TimeGenerated descSPL
sourcetype="openattacks:audit"
EventType IN ("agent.config.write", "agent.approval.change")
| table _time EventType Actor Asset Action Target CorrelationId
| sort - _timeInvestigation steps
- Diff command, args, environment, origin, autoApprove and task settings against the previously approved version.
- Correlate the author and change with the next agent launch; inspect path canonicalization and trailing-dot variants where relevant.
02OAuth discovery supplies an attacker URLUnexpected agent tool useCustom schema · Draft
A 401 response leads to an attacker-controlled authorization endpoint.
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.
03URL enters PowerShell through open()Suspicious Windows script executionNative logs · Draft
Affected Windows or WSL browser-opening code fails to escape the target.
What to look for
A script host or shell runs encoded content, a remote URL, or inline script syntax, then downloads or launches a payload.
Visibility: Defender-side logs where collected
Required logs
- Windows process creation: Sysmon Event 1 or Defender DeviceProcessEvents
- PowerShell script-block logging when enabled
- Download / network and installer events
Coverage limit. Encoded commands and remote administration can be legitimate. This does not cover every script host, in-process execution, or non-Windows branch.
Windows process candidates. Sigma uses normalized process_creation fields; KQL requires Defender for Endpoint; SPL requires the Sysmon XML sourcetype. Tune collectors, fields and baselines before use.
Sigma
title: Hunt - Suspicious Windows script execution
id: 2bb88802-9f03-4d99-a192-fe44c9484e8b
status: experimental
description: A script host or shell runs encoded content, a remote URL, or
inline script syntax, then downloads or launches a payload.
author: MagicSword OpenAttacks catalog
date: 2026-10-23
falsepositives:
- Authorized activity matching this broad hunt; baseline and investigate
before alerting.
level: low
logsource:
category: process_creation
product: windows
detection:
selection:
Image|endswith:
- \powershell.exe
- \pwsh.exe
- \mshta.exe
- \wscript.exe
- \cscript.exe
CommandLine|contains:
- -enc
- http
- FromBase64String
- "javascript:"
- "vbscript:"
condition: selection
KQL
// Draft hunt; requires Defender for Endpoint process telemetry for this OS.
DeviceProcessEvents
| where Timestamp > ago(7d)
| where FileName in~ ("powershell.exe", "pwsh.exe", "mshta.exe", "wscript.exe", "cscript.exe")
| where ProcessCommandLine contains "-enc" or ProcessCommandLine contains "http" or ProcessCommandLine contains "FromBase64String" or ProcessCommandLine contains "javascript:" or ProcessCommandLine contains "vbscript:"
| project Timestamp, DeviceName, AccountName, FileName, ProcessCommandLine, InitiatingProcessFileName, InitiatingProcessCommandLine, SHA1
| order by Timestamp descSPL
sourcetype="XmlWinEventLog:Microsoft-Windows-Sysmon/Operational" EventCode=1
Image IN ("*\\powershell.exe", "*\\pwsh.exe", "*\\mshta.exe", "*\\wscript.exe", "*\\cscript.exe")
CommandLine IN ("*-enc*", "*http*", "*FromBase64String*", "*javascript:*", "*vbscript:*")
| table _time host User Image CommandLine ParentImage ParentCommandLine
| sort - _timeInvestigation steps
- Follow the parent and child process tree and recover the script or URL from logs without executing it.
- Correlate the payload's signer/hash, persistence, and network activity; baseline approved automation.
04Commands execute before the browser opensSuspicious Windows script executionNative logs · Draft
The injection occurs in the URL-opening path without requiring prompt injection.
What to look for
A script host or shell runs encoded content, a remote URL, or inline script syntax, then downloads or launches a payload.
Visibility: Defender-side logs where collected
Required logs
- Windows process creation: Sysmon Event 1 or Defender DeviceProcessEvents
- PowerShell script-block logging when enabled
- Download / network and installer events
Coverage limit. Encoded commands and remote administration can be legitimate. This does not cover every script host, in-process execution, or non-Windows branch.
Windows process candidates. Sigma uses normalized process_creation fields; KQL requires Defender for Endpoint; SPL requires the Sysmon XML sourcetype. Tune collectors, fields and baselines before use.
Sigma
title: Hunt - Suspicious Windows script execution
id: 2bb88802-9f03-4d99-a192-fe44c9484e8b
status: experimental
description: A script host or shell runs encoded content, a remote URL, or
inline script syntax, then downloads or launches a payload.
author: MagicSword OpenAttacks catalog
date: 2026-10-23
falsepositives:
- Authorized activity matching this broad hunt; baseline and investigate
before alerting.
level: low
logsource:
category: process_creation
product: windows
detection:
selection:
Image|endswith:
- \powershell.exe
- \pwsh.exe
- \mshta.exe
- \wscript.exe
- \cscript.exe
CommandLine|contains:
- -enc
- http
- FromBase64String
- "javascript:"
- "vbscript:"
condition: selection
KQL
// Draft hunt; requires Defender for Endpoint process telemetry for this OS.
DeviceProcessEvents
| where Timestamp > ago(7d)
| where FileName in~ ("powershell.exe", "pwsh.exe", "mshta.exe", "wscript.exe", "cscript.exe")
| where ProcessCommandLine contains "-enc" or ProcessCommandLine contains "http" or ProcessCommandLine contains "FromBase64String" or ProcessCommandLine contains "javascript:" or ProcessCommandLine contains "vbscript:"
| project Timestamp, DeviceName, AccountName, FileName, ProcessCommandLine, InitiatingProcessFileName, InitiatingProcessCommandLine, SHA1
| order by Timestamp descSPL
sourcetype="XmlWinEventLog:Microsoft-Windows-Sysmon/Operational" EventCode=1
Image IN ("*\\powershell.exe", "*\\pwsh.exe", "*\\mshta.exe", "*\\wscript.exe", "*\\cscript.exe")
CommandLine IN ("*-enc*", "*http*", "*FromBase64String*", "*javascript:*", "*vbscript:*")
| table _time host User Image CommandLine ParentImage ParentCommandLine
| sort - _timeInvestigation steps
- Follow the parent and child process tree and recover the script or URL from logs without executing it.
- Correlate the payload's signer/hash, persistence, and network activity; baseline approved automation.
Technical detailsTooling, impact
Attack context
- AI's role
- Agent / MCP as target
- Initial access
- Developer connects coding agent to attacker-controlled or later-compromised remote MCP server
Tooling as reported
AI products, models and artifacts
- opencode / opencode-ai 1.0.137–1.18.32
- MCP OAuth discovery
Other tooling
- npm open@10.1.2
- PowerShell
- related class CVEs: CVE-2025-6514 mcp-remote, CVE-2025-58062 openmcp-client, CVE-2025-10619 sequa-mcp
Targets and reported impact
- Targets / sectors
- Windows and WSL developers using opencode (~3M weekly npm downloads cited by Pillar)
- Reported scale / exposure
- Fixed in 1.18.33 (2026-09-28) without CVE/advisory; proposed CVSS 9.8
Reported attribution
Vulnerability research (no in-the-wild campaign named)
Structured attribution
- Actor
- Vulnerability research (no in-the-wild campaign named)
Detection rules
- Review attack-specific indicatorsGUIDANCE
Monitor for the techniques described in opencode remote MCP OAuth RCE. Refer to sources for specific IOCs and TTPs.
Source defensive notes
Upgrade opencode; inventory installations (no CVE for scanners); pull unvetted remote MCP; argv not shell for URLs; MagicSword prevent unapproved second-stage binaries
Sources & research credit
1 sourceResearch by Eilon Cohen · Pillar Security
Record & downloads
- Source record name
- opencode remote MCP OAuth RCE
- Entry ID
- opencode-mcp-oauth-rce-2026-10
- Catalog review
- October 20, 2026
- Category
- attacks-on-ai
- AI involvement confidence
- confirmed