Aruba Iap Publish

• Outcome: Review
• Latest audit: May 28, 2026 · 6:18 AM UTC
• Version: 1.2.0

Overview

This appears to be a real Aruba access-point management skill, but it needs review because it can change live network devices and handles credentials and local installation unsafely.

Install only after reviewing the code and using it in a controlled admin environment. Prefer SSH keys, avoid command-line passwords, do not use the default helper-script credentials, keep backups and monitor outputs in a protected directory, and treat apply/rollback operations as live network changes that require authorization and a tested recovery plan.

SkillSpector

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep
• Behavioral AST: exec() Call, eval() Call, Dynamic Import
• MCP Least Privilege: Underdeclared Capability, Wildcard Permission, Missing Permission Declaration
• MCP Tool Poisoning: Hidden Instructions, Unicode Deception, Parameter Description Injection

Findings (34)

subprocess module call

Medium

• Category: Dangerous Code Execution

Content

"""Execute command on IAP via SSH"""
        try:
            cmd = f'echo "{self.password}" | sshpass -p stdin ssh -o StrictHostKeyChecking=no -o ConnectTimeout=30 {self.username}@{self.ap_ip} "{command}"'
            result = subprocess.run(cmd, shell=True, capture_output=True, text=True, timeout=60)
            return result.stdout
        except subprocess.TimeoutExpired:
            return "ERROR: Connection timeout"

• Confidence: 99% confidence
• Finding: The script builds a shell command from untrusted values including username, password, AP IP, and command, then executes it with shell=True. This creates a command injection path and also exposes credentials through shell invocation and process arguments, making compromise of the host running the backup plausible.

subprocess module call

Medium

• Category: Dangerous Code Execution

Content

try:
            # Use sshpass for password authentication if available
            cmd = f'echo "{self.password}" | sshpass -p stdin ssh {self.username}@{self.ap_ip} "{command}"'
            result = subprocess.run(cmd, shell=True, capture_output=True, text=True, timeout=30)
            return result.stdout
        except subprocess.TimeoutExpired:
            return "ERROR: Connection timeout"

• Confidence: 99% confidence
• Finding: The code builds a shell command from untrusted values (username, password, ap_ip, and command) and executes it with shell=True. An attacker controlling any of those values can inject arbitrary shell syntax and achieve local command execution, and the password is also exposed via insecure shell piping.

subprocess module call

Medium

• Category: Dangerous Code Execution

Content

"""Execute command on IAP via SSH"""
        try:
            cmd = f'echo "{self.password}" | sshpass -p stdin ssh -o StrictHostKeyChecking=no -o ConnectTimeout=10 {self.username}@{self.ap_ip} "{command}"'
            result = subprocess.run(cmd, shell=True, capture_output=True, text=True, timeout=15)
            return result.stdout
        except Exception as e:
            return ""

• Confidence: 98% confidence
• Finding: This is a true vulnerability because the script builds a shell command from untrusted inputs (`username`, `ap_ip`, `password`, and `command`) and executes it with `shell=True`. That enables command injection on the local system running the script, and the surrounding use of plaintext credentials and disabled host key checking increases the likelihood of credential theft or unauthorized command execution.

Lp3

Medium

• Category: MCP Least Privilege
• Confidence: 91% confidence
• Finding: The skill documentation advertises capabilities that imply environment access, file read/write, and shell execution, but it does not declare corresponding permissions. In an agent ecosystem, undeclared capabilities undermine least-privilege controls and make it harder for operators to assess what the skill can actually do, increasing the chance of unintended command execution or filesystem modification.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 97% confidence
• Finding: The documented purpose frames the skill as a managed Aruba IAP configuration tool, but the broader behavior includes direct device access via helper scripts, insecure default credentials, and installation behavior that writes executables into system paths. This mismatch is dangerous because users may trust and run the skill for routine network management without realizing it can perform privileged local changes and connect to devices in less controlled ways, magnifying the risk of credential compromise and unauthorized system modification.

Intent-Code Divergence

Medium

• Confidence: 96% confidence
• Finding: The documented rollback for syslog level repeats the forward commands instead of restoring the prior state, so operators relying on this guidance may believe they have a recovery path when they do not. In a network configuration management skill, incorrect rollback instructions can prolong outages, leave logging misconfigured during incident response, and undermine safe-change guarantees advertised by the tool.

Intent-Code Divergence

Medium

• Confidence: 96% confidence
• Finding: The verify() function records expected-state verification as a passing check even though the comparison logic is explicitly not implemented. This can create false assurance after a change, causing operators or downstream automation to trust a validation result that never actually verified the intended state.

Context-Inappropriate Capability

Medium

• Confidence: 93% confidence
• Finding: The secret resolver accepts `env:` and `file:` references and will read any environment variable or local file path supplied by the caller. In an agent skill context, this creates an overly broad secret access surface unrelated to Aruba IAP management and can expose host secrets, API keys, SSH keys, or other sensitive local data if untrusted input reaches `get_secret`.

Description-Behavior Mismatch

Medium

• Confidence: 84% confidence
• Finding: The script goes beyond configuration backup and also collects client-device inventory, which may include sensitive network metadata about connected users or endpoints. In an infrastructure-management skill this is not inherently malicious, but it increases data exposure and privacy risk if operators do not expect or protect these files.

Context-Inappropriate Capability

Medium

• Confidence: 98% confidence
• Finding: The script embeds default credentials (aruba/aruba123) and uses them automatically for SSH access. Hardcoded default credentials are dangerous because they normalize insecure deployments, may work against devices left at defaults, and combine with shell execution to create unauthorized access and credential exposure risks.

Context-Inappropriate Capability

Medium

• Confidence: 97% confidence
• Finding: The script explicitly disables SSH host key verification (`StrictHostKeyChecking=no`) and uses shell-based password injection via `sshpass`, which weakens transport security for a monitoring task. In this context, that makes man-in-the-middle attacks and credential interception materially easier, especially since the tool is intended to connect to network infrastructure devices.

Context-Inappropriate Capability

Medium

• Confidence: 98% confidence
• Finding: The script accepts an SSH password as a positional command-line argument and appends it directly to the invoked command. Command-line arguments are commonly exposed through process listings, shell history, audit logs, and CI/job telemetry, so this can leak credentials to other local users or monitoring systems. In a network-management skill, this is more dangerous because the credential may grant administrative access to wireless infrastructure.

Missing User Warnings

Medium

• Confidence: 89% confidence
• Finding: The README provides examples for storing secrets in a local `secrets.json` file and exporting credentials via environment variables, but it does not warn users against committing plaintext secret files to version control, leaving them in example directories, or exposing them through shell history and process environments. In a network-management skill that handles RADIUS and WPA credentials, this omission can lead to credential leakage and unauthorized access to Aruba infrastructure or dependent authentication systems.

Missing User Warnings

Medium

• Confidence: 85% confidence
• Finding: The documentation provides plain-text secret examples in files and shell exports without clearly warning that secrets may be exposed through shell history, process inspection, terminal logs, screenshots, or accidental commits. In an infrastructure-management context handling Wi-Fi PSKs and RADIUS keys, this can directly lead to credential disclosure and compromise of network access controls.

Missing User Warnings

Medium

• Confidence: 92% confidence
• Finding: The quick reference includes plaintext example secrets such as SNMP community strings, WPA passphrases, and environment/file-based secret values without a strong warning that these are illustrative only and must not be reused. In operator-facing documentation, this can normalize insecure handling of credentials, leading users to paste real secrets into files, shell history, terminal logs, or version control.

Missing User Warnings

Medium

• Confidence: 84% confidence
• Finding: The command table labels `apply-cmd` and `rollback-cmd` with risk levels, but it does not clearly warn that these operations can be disruptive and may affect live network connectivity. Because this skill manages Aruba IAP configuration, users may treat the quick reference as operationally safe guidance and execute high-impact changes without maintenance planning or prechecks.

Missing User Warnings

Medium

• Confidence: 94% confidence
• Finding: This documentation includes example secret values and explicitly instructs users to print or view secrets via `echo $RADIUS_PRIMARY_KEY` and `cat secrets.json`, which can expose credentials in terminal history, screen recordings, shared shells, logs, or copied documentation snippets. In a network configuration skill, these secrets may control wireless authentication or RADIUS access, so disclosure could enable unauthorized access or lateral movement.

Missing User Warnings

Medium

• Confidence: 93% confidence
• Finding: The README includes a CLI example that passes `--ssh-password yourpassword` directly on the command line. This can expose credentials through shell history, process listings, terminal logging, and copy/paste into tickets or chat, making accidental disclosure more likely. In the context of a network-management tool for Aruba IAPs, disclosed admin credentials could enable unauthorized access to wireless infrastructure.

Missing User Warnings

Medium

• Confidence: 97% confidence
• Finding: This code saves the generated command set directly to `commands.json` and `commands.txt`, while `generate_commands()` may have already resolved `secret_ref` values into plaintext RADIUS/auth keys and WPA passphrases. Because there is no warning, redaction-before-write, or protected storage behavior, sensitive credentials can be unintentionally persisted to disk where they may be exposed through backups, logs, shared workspaces, or local compromise.

Missing User Warnings

Medium

• Confidence: 89% confidence
• Finding: The monitor routine collects and persists raw outputs from commands such as "show configuration", client tables, authentication traces, logs, routing, and SNMP settings directly to disk. These artifacts can contain credentials, network topology, client identifiers, security events, and other sensitive operational data, so writing them without explicit consent controls, redaction, or protective storage measures creates a real confidentiality risk if the output directory is accessible, backed up, or exfiltrated.

Missing User Warnings

Medium

• Confidence: 87% confidence
• Finding: The snapshot() workflow writes full running configuration and user-table output to disk, which may include SSIDs, usernames, network topology, and other sensitive operational data. If the output directory is shared, weakly protected, or collected by other systems, this becomes an information exposure risk.

Missing User Warnings

Medium

• Confidence: 88% confidence
• Finding: The `env:` path reads arbitrary environment variables silently, with no warning, consent, or audit trail. In a skill running on a user's workstation or automation host, this can unexpectedly pull sensitive process secrets that have nothing to do with AP configuration, making misuse or accidental overreach harder to detect.

Missing User Warnings

Medium

• Confidence: 90% confidence
• Finding: The `file:` path reads arbitrary local files with no user-facing warning or approval, enabling the skill to access any readable path provided to it. In this context, that could expose unrelated local credentials, configuration files, or private keys from the host system, which is broader than the stated network device management purpose.

Missing User Warnings

Medium

• Confidence: 96% confidence
• Finding: The installer unconditionally writes an executable wrapper and symlink into /opt/homebrew, modifying a global executable path without prompting the user, checking permissions, or warning about system-wide effects. This is dangerous because installation side effects on shared system paths can overwrite existing tools, require elevated privileges, and make unintended persistence or path hijacking harder for users to notice or control.

Missing User Warnings

Medium

• Confidence: 95% confidence
• Finding: The document includes `write erase startup-config` and `reload` as a factory reset workflow without an explicit warning that these commands are destructive and can permanently remove device configuration and interrupt network service. In a configuration-management skill for Aruba IAPs, operators may copy commands directly, so omission of a prominent caution materially increases the risk of accidental outages or loss of access.

VirusTotal

66/66 vendors flagged this skill as clean.

Static analysis

No suspicious patterns detected.