FakeGit agentbaiting - 800+ fake AI skills and MCP servers recommended by agents
Oct 30, 2026: Island Security identified 'FakeGit' campaign: 800+ fake GitHub repositories impersonating popular AI skills, MCP servers, Cursor rules, and Claude Code extensions. Repos use…
What happened
Oct 30, 2026: Island Security identified 'FakeGit' campaign: 800+ fake GitHub repositories impersonating popular AI skills, MCP servers, Cursor rules, and Claude Code extensions. Repos use typosquat/combosquat names (e.g., cursor-rules-best-practices, mcp-server-github-enhanced, claude-code-skills-security), contain trojanized npm/PyPI packages or malicious install scripts. Adversary optimized repo metadata (README, topics, stars, forks) to rank highly in LLM tool recommendations: when users ask agents find me a GitHub MCP server or what Cursor skills should I use?, agents recommend FakeGit repos from search results. Payloads: Shai-Hulud variants, Discord webhook exfiltration, npm/PyPI worms, SSH backdoors. Island tracked 2,400+ installations from agent recommendations Oct-Nov 2026. Island reports the repositories deliver SmartLoader, which establishes persistence and installs StealC to steal credentials, active sessions, and other sensitive data. Attack exploits agent trust in GitHub search ranking and lack of package verification.
Adversary creates 800+ fake GitHub repos mimicking legitimate AI skills, MCP servers, and Cursor rules; optimizes for LLM recommendation via SEO and naming; agents recommend malicious packages to users
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 · Create fake repos
Adversary creates 800+ GitHub repos with typosquat/combosquat names mimicking popular AI skills, MCP servers, Cursor rules; optimizes metadata (README, topics, keywords) for LLM search ranking; inflates stars/forks
Hunt: Related accounts or assistants repeat correlated posting, forwarding or tool actions beyond the expected user workflow.
Requires platform-side visibility. Similar content and synchronized activity can have legitimate explanations.
Source for step 1 - 02 · Agent recommends malicious repo
User asks AI agent to find/recommend skill, MCP server, or Cursor rule; agent searches GitHub/web, ranks FakeGit repos highly due to optimized metadata; agent recommends malicious repo to user as legitimate tool
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 · User installs per agent recommendation
User trusts agent recommendation and clones repo or installs package; malicious install script or npm/PyPI postinstall executes; Shai-Hulud stealer, SSH backdoor, or worm payload deploys
Hunt: A package-manager install or lifecycle script launches a shell, downloader, or script interpreter that retrieves or executes an unexpected second stage.
Build and install scripts commonly invoke shells and download dependencies. Require package provenance and destination context; this hunt does not by itself prove supply-chain compromise.
Source for step 3 - 04 · Credential theft and persistence
Payload harvests AI API keys, GitHub tokens, SSH keys, AWS credentials; establishes persistence (SSH backdoor, cron job); exfiltrates to Discord webhooks or C2; 2,400+ successful compromises tracked Oct-Nov 2026
Hunt: An unexpected process or tool accesses credential stores and is followed by token use or outbound transfer inconsistent with the owner.
Defender DeviceFileEvents is not a general file-read audit trail. Collect suitable object-access, EDR or application logs; legitimate developer tools also read secrets.
Source for step 4
Why it matters
Reported entry point: AI agent recommends malicious GitHub repo in response to user query; user clones/installs per agent suggestion. Reported impact: Payload harvests AI API keys, GitHub tokens, SSH keys, AWS credentials; establishes persistence (SSH backdoor, cron job); exfiltrates to Discord webhooks or C2; 2,400+ successful compromises tracked Oct-Nov 2026; reported targets include 2,400+ installations tracked Oct-Nov 2026 from agent recommendations; AI developers using agent-assisted package discovery; 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
Verify GitHub repos before install (check org, stars, age, commit history); use allowlisted skill/MCP sources; implement agent guardrails to warn on unverified packages; educate users not to blindly trust agent recommendations; review install scripts; use SCA tools; prefer official sources (Anthropic MCP registry, Cursor skills directory); flag typosquat patterns
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.
01Create fake reposCoordinated messaging / propagationCustom schema · Draft
Adversary creates 800+ GitHub repos with typosquat/combosquat names mimicking popular AI skills, MCP servers, Cursor rules; optimizes metadata (README, topics, keywords) for LLM search ranking; inflates stars/forks
What to look for
Related accounts or assistants repeat correlated posting, forwarding or tool actions beyond the expected user workflow.
Visibility: Platform / application records
Required logs
- Platform account and posting audit
- Agent session and message-provenance records
- Rate, recipient and content-similarity features
Coverage limit. Requires platform-side visibility. Similar content and synchronized activity can have legitimate explanations.
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 - Coordinated messaging / propagation
id: c4001568-b68b-4887-a86f-6f05807640f6
status: experimental
description: Related accounts or assistants repeat correlated posting,
forwarding or tool actions beyond the expected user workflow.
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:
- social.post
- message.forward
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 ("social.post", "message.forward")
| project TimeGenerated, EventType, Actor, Asset, Action, Target, CorrelationId
| order by TimeGenerated descSPL
sourcetype="openattacks:audit"
EventType IN ("social.post", "message.forward")
| table _time EventType Actor Asset Action Target CorrelationId
| sort - _timeInvestigation steps
- Group actions by linked account, session, recipient and time rather than judging writing style.
- Compare propagation with legitimate campaigns and preserve whether the source describes a lab demonstration or observed abuse.
02Agent recommends malicious repoUntrusted content enters an agent sessionCustom schema · Draft
User asks AI agent to find/recommend skill, MCP server, or Cursor rule; agent searches GitHub/web, ranks FakeGit repos highly due to optimized metadata; agent recommends malicious repo to user as legitimate tool
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.
Review recommendation provenance and the subsequent action. A repository appearing in search results is not compromise.
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.
03User installs per agent recommendationPackage installation spawns a shell or downloaderNative logs · Draft
User trusts agent recommendation and clones repo or installs package; malicious install script or npm/PyPI postinstall executes; Shai-Hulud stealer, SSH backdoor, or worm payload deploys
What to look for
A package-manager install or lifecycle script launches a shell, downloader, or script interpreter that retrieves or executes an unexpected second stage.
Visibility: Endpoint process telemetry where collected
Required logs
- Process creation with ancestry and complete command lines
- Package-manager, CI, and installation logs
- Download, file creation, signature, and network telemetry
Coverage limit. Build and install scripts commonly invoke shells and download dependencies. Require package provenance and destination context; this hunt does not by itself prove supply-chain compromise.
Process candidates with package-manager ancestry that launch shells or downloaders. Tune known build tooling and correlate with network/file telemetry.
MCP reference matches: Curl Download Activity from npm Package Install (Elastic detection-rules), Suspicious PowerShell from npm Package Install (Elastic detection-rules). The query below is a catalog adaptation, not a verbatim upstream rule.
Sigma
title: Hunt - Shell or Downloader Spawned by Package Installation
id: c83f22aa-d07c-4f3c-a67f-6d5512818133
status: experimental
description: Detects package-manager processes spawning common shells, interpreters, or downloaders.
author: MagicSword OpenAttacks catalog
date: 2026-11-06
references:
- https://github.com/elastic/detection-rules/blob/main/rules/cross-platform/command_and_control_curl_download_from_npm_install.toml
falsepositives:
- Legitimate package build and installation scripts
level: medium
logsource:
category: process_creation
detection:
selection_parent:
ParentCommandLine|contains:
- npm-cli.js
- npx-cli.js
- yarn
- pnpm
- pip install
- python -m pip
- gem install
selection_child:
Image|endswith:
- \powershell.exe
- \pwsh.exe
- /sh
- /bash
- /zsh
- /curl
- /wget
- /python
- /node
condition: all of selection_*
KQL
DeviceProcessEvents
| where Timestamp > ago(7d)
| where InitiatingProcessCommandLine has_any ("npm-cli.js", "npx-cli.js", "yarn", "pnpm",
"pip install", "python -m pip", "gem install")
| where FileName in~ ("powershell.exe", "pwsh.exe", "sh", "bash", "zsh", "curl", "wget", "python", "node")
| project Timestamp, DeviceName, AccountName, FileName, ProcessCommandLine,
InitiatingProcessFileName, InitiatingProcessCommandLine, SHA1
| order by Timestamp descSPL
| tstats count min(_time) as firstTime max(_time) as lastTime
from datamodel=Endpoint.Processes
where Processes.parent_process IN ("*npm-cli.js*", "*npx-cli.js*", "*yarn*", "*pnpm*",
"*pip install*", "*python -m pip*", "*gem install*")
Processes.process_name IN ("powershell.exe", "pwsh.exe", "sh", "bash", "zsh", "curl", "wget", "python", "node")
by Processes.dest Processes.user Processes.process_name Processes.process
Processes.parent_process_name Processes.parent_process
| `drop_dm_object_name(Processes)`
| sort - lastTimeInvestigation steps
- Identify the package, version, publisher, lifecycle hook, resolved source, and parent process.
- Recover downloaded content and correlate credential access, persistence, or outbound traffic.
04Credential theft and persistenceSensitive store access → credential useCustom schema · Draft
Payload harvests AI API keys, GitHub tokens, SSH keys, AWS credentials; establishes persistence (SSH backdoor, cron job); exfiltrates to Discord webhooks or C2; 2,400+ successful compromises tracked Oct-Nov 2026
What to look for
An unexpected process or tool accesses credential stores and is followed by token use or outbound transfer inconsistent with the owner.
Visibility: Defender-side logs where collected
Required logs
- File-read / API access telemetry for relevant secret stores
- Process identity and command lines
- Identity, token and outbound-network audit logs
Coverage limit. Defender DeviceFileEvents is not a general file-read audit trail. Collect suitable object-access, EDR or application logs; legitimate developer tools also read secrets.
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 - Sensitive store access → credential use
id: 46c46c62-2d83-4198-a8bf-d6637094527b
status: experimental
description: An unexpected process or tool accesses credential stores and is
followed by token use or outbound transfer inconsistent with the owner.
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:
- secret.read
- credential.store.read
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 ("secret.read", "credential.store.read")
| project TimeGenerated, EventType, Actor, Asset, Action, Target, CorrelationId
| order by TimeGenerated descSPL
sourcetype="openattacks:audit"
EventType IN ("secret.read", "credential.store.read")
| table _time EventType Actor Asset Action Target CorrelationId
| sort - _timeInvestigation steps
- Identify the actor, process and exact store touched, then pivot to token use from new devices or infrastructure.
- Correlate reads with egress and rotate affected credentials after confirming exposure.
Technical detailsTooling, impact & MITRE ATLAS
Attack context
- AI's role
- Agent / MCP as target
- Additional role
- AI tool as lure
- Initial access
- AI agent recommends malicious GitHub repo in response to user query; user clones/installs per agent suggestion
Tooling as reported
AI products, models and artifacts
- Cursor IDE and Cursor rules
- Claude Code
- MCP (Model Context Protocol) servers
- AI coding assistants (generic)
- LLM search and recommendation systems
Other tooling
- Fake GitHub repositories (800+)
- npm/PyPI trojanized packages
- Shai-Hulud credential stealer variants
- SSH backdoors
- Discord webhook exfiltration
- SmartLoader dropper
- StealC infostealer
Targets and reported impact
- Targets / sectors
- 2,400+ installations tracked Oct-Nov 2026 from agent recommendations; AI developers using agent-assisted package discovery
- Reported scale / exposure
- 800+ fake GitHub repos created; 2,400+ installations from agent recommendations (Oct-Nov 2026); ongoing campaign as of Nov 2026
- Target geography
- Multiple regions
Reported attribution
Unknown actor; 'FakeGit' campaign; sophisticated understanding of LLM recommendation mechanics and AI developer workflows
Structured attribution
- Actor
- FakeGit
- Country
- Unknown
- Motivation
- financially-motivated
- Attributed by
- Island
- Attribution confidence
- high
Detection rules
- Potential Browser Data StealingSIGMAView rule
Detects StealC infostealer browser data theft on Windows
- Suspicious File Access to Browser Credential StorageSIGMAView rule
Detects SmartLoader/StealC accessing browser credential files
- Malicious skill and MCP repository detectionGUIDANCE
Monitor agent-initiated GitHub clones; alert on typosquat repo names; verify repo authenticity (check org, stars, age, commit history); detect inflated stars/forks with low commit activity; scan for SmartLoader/StealC signatures
Activity timeline
- First seen
- 2026-10-30
- Last updated
- 2026-11-04
Updates
- 2026-10-30 — Island
Initial discovery and analysis
Source defensive notes
Verify GitHub repos before install (check org, stars, age, commit history); use allowlisted skill/MCP sources; implement agent guardrails to warn on unverified packages; educate users not to blindly trust agent recommendations; review install scripts; use SCA tools; prefer official sources (Anthropic MCP registry, Cursor skills directory); flag typosquat patterns
MITRE ATLAS
AI-specific catalog annotations based on the reported behavior; these are not MITRE case-study assignments.
Adversary manipulates LLM tool recommendation systems to deliver malicious packages via agent-suggested GitHub repos Source
Sources & research credit
1 sourceResearch by Island Security Research · Island
Island Security analysis is the live source for the FakeGit SmartLoader and StealC chain. The previously cited SecurityBreak post returned HTTP 404 and was removed. First documented campaign explicitly optimizing malware for LLM recommendation engines (agentbaiting). Represents new supply chain threat vector where adversaries target agent search/recommendation behavior rather than direct user search. Detection: monitor agent-initiated GitHub clones, alert on typosquat repo names (edit distance to popular repos), check for new GitHub accounts with multiple similar repos, review repos with inflated stars/forks but low commit activity, flag repos created after popular tools rise, scan for SmartLoader/StealC signatures.
Record & downloads
- Source record name
- FakeGit agentbaiting - 800+ fake AI skills and MCP servers recommended by agents
- Entry ID
- fakegit-agentbaiting-island-2026-10
- Catalog review
- November 10, 2026
- Category
- attacks-on-ai
- AI involvement confidence
- confirmed