Back to skill

Security audit

Cluster Gateway

Security checks for vulnerabilities and agentic risk

Overview

The skill is a cluster gateway guide, but it documents high-impact remote orchestration and an all-interface gateway bind without clear access-control or safety boundaries.

Review this carefully before installing or using it against a real cluster. Only use the gateway on trusted networks with explicit authentication, encrypted transport, least-privilege service accounts, command validation, logging, and approval controls for deploy, task, and relay operations.

Vulnerability Patterns
  • Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
Findings (1)

T09 · Insecure Skill Coding Practices

Warning
Location
SKILL.md:15
Finding

Network-Exposed Cluster Gateway Lacks Documented Access Controls

Content
View full analysis
--agent # Task an Worker Node senden openclaw cluster gateway task --target --command "" # Relay Node konfigurieren openclaw cluster gateway relay --enable --node ``` ```json // cluster-gateway.config.json { "gateway": { "id": "node-1", "role": "hub", "bind": "0.0.0.0", "port": 18789 }, "nodes": [ {"id": "node-2", "role": "worker", "host": "10.10.0.2"}, {"id": "node-3", "role": "relay", "host": "10.10.0.3"}, {"id": "node-4", "role": "worker", "host": "10.10.0.4"} ], "messaging": { "enabled": false, "provider": "pending" } } ``` ### Technical Analysis The documented gateway binds to `0.0.0.0`, making it reachable through every network interface unless an external firewall restricts access. The same gateway provides security-sensitive orchestration operations, including deploying sub-agents, enabling relay nodes, and sending task strings through the `--command` argument. The documentation does not specify authentication, authorization, transport encryption, per-node identity verification, command allowlisting, or network-level restrictions. Accepting unrestricted task strings is particularly dangerous if the implementation forwards them to a command interpreter or otherwise executes them without strict validation. Because the reviewed project contains only documentation and no gateway implementation, the presence or absence of controls in the underlying `openclaw` service cannot be verified. The vulnerability applies if the documented configuration is deployed without protections supplied elsewhere. ### Att ...[truncated 1526 chars]
Remediation
View remediation
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (2)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The skill documents commands for deploying agents, sending remote tasks, and enabling relay behavior across cluster nodes, but provides no warnings, authorization requirements, confirmation steps, or safety boundaries. In an orchestration context, these actions can change remote system state or trigger command execution on other nodes, making accidental misuse or unsafe automation more likely.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
98% confidence
Finding

The natural-language instructions and descriptions are presented in German throughout the file, and there is no indication that users may select another language or that the skill is intentionally restricted to a German-speaking context. This matches the policy pattern of forcing a specific language without user opt-in.

Content

No source excerpt is available for this finding.