ClawHub
Back to skill

Security audit

MailboxValidator

Security checks for vulnerabilities and agentic risk

Overview

The skill is mostly a coherent MailboxValidator connector, but its setup instructions tell users or agents to execute remote installer scripts directly, which needs review before installation.

Install only if you are comfortable using OOMOL as an intermediary for MailboxValidator requests. Do not run the pasted installer commands blindly; prefer a verified, versioned oo CLI install method and confirm any authentication or account-connection steps yourself.

A.I.G

Vulnerability Patterns
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • 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
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Findings (2)

T03 · Remote Payload Retrieval and Execution

Error
Location
SKILL.md:57
Finding

Unverified Remote Installer Executed Directly by Shell

Content
View full analysis

Vulnerability Details

File Location: SKILL.md:57
Vulnerability Type: Remote payload retrieval and execution
Risk Level: Critical

Vulnerable Code:

bash
curl -fsSL https://cli.oomol.com/install.sh | bash    # macOS / Linux

Technical Analysis

The installation instruction downloads a mutable remote script and sends it directly to Bash. The package does not pin the installer to a reviewed version, verify a cryptographic signature or checksum, or save the script for inspection before execution.

HTTPS protects the connection in transit but does not establish that the script is immutable or safe. Compromise of the hosting infrastructure, DNS or certificate ecosystem, publishing account, or installer build process could cause different code from the reviewed content to execute. Direct piping also creates inspection and partial-download hazards because the shell processes content as it arrives.

Installing the CLI may be relevant to the Skill's operation, but executing an unverified remote payload is not the minimum capability required. A pinned and independently verified installation mechanism would achieve the same purpose with less risk.

Attack Path

  1. The oo CLI is unavailable, and an agent or user follows the first-time setup instructions.
  2. An attacker compromises or gains control over the remote installer delivery path.
  3. The response from https://cli.oomol.com/install.sh contains attacker-controlled shell commands.
  4. curl streams that response directly into Bash without integrity verification or review.
  5. Bash executes the attacker's commands with the privileges of the user running the installation.

Impact Assessment

Arbitrary command execution is possible within the invoking user's security context. Depending on that user's permissions and environment, a malicious installer could read or alter files, access locally available credentials and tokens, install additional ...[truncated 271 chars]

Remediation
View remediation

Remediation Suggestions

  • Remove the curl | bash installation pattern.
  • Prefer a trusted operating-system package manager and pin the CLI to a reviewed version.
  • Alternatively, download a versioned release artifact without executing it immediately.
  • Publish and verify a cryptographic signature from a separately managed trusted key. If signatures are unavailable, verify a pinned SHA-256 checksum obtained through an independent trusted channel.
  • Inspect the verified installer before execution and run it with the least-privileged account necessary.
  • Document the expected files, permissions, and system changes made by installation.
  • Do not automatically install software when an action fails; return an actionable error and require informed user approval.

T03 · Remote Payload Retrieval and Execution

Error
Location
SKILL.md:61
Finding

Unverified Remote PowerShell Installer Executed Directly

Content
View full analysis

Vulnerability Details

File Location: SKILL.md:61
Vulnerability Type: Remote payload retrieval and execution
Risk Level: Critical

Vulnerable Code:

powershell
irm https://cli.oomol.com/install.ps1 | iex           # Windows PowerShell

Technical Analysis

This instruction retrieves a mutable PowerShell script using Invoke-RestMethod (irm) and immediately executes the returned content using Invoke-Expression (iex). No version pinning, signature validation, checksum verification, or local inspection occurs before execution.

The effective payload is controlled by the remote response and can change after the Skill has been audited. Although HTTPS protects transport confidentiality and integrity under normal conditions, it does not mitigate compromise of the publisher, hosting platform, release process, or trusted delivery infrastructure.

Installing the CLI may support the declared functionality, but immediate execution of unverified network content exceeds the minimum privileges and trust required for installation.

Attack Path

  1. The oo CLI is missing on a Windows host, and the setup instruction is followed.
  2. An attacker compromises the installer host, publishing process, account, or another trusted component of the delivery path.
  3. Invoke-RestMethod downloads attacker-controlled PowerShell source.
  4. The pipeline passes that source directly to Invoke-Expression.
  5. PowerShell executes the payload with the invoking user's permissions.

Impact Assessment

The attacker could execute arbitrary PowerShell operations in the user's context, including reading or modifying accessible files, extracting locally available credentials, downloading further payloads, changing user-level configuration, or creating persistence. Administrator-level impact is possible only if the command is run from an elevated PowerShell session or the payload separately succeeds in escalating privilege ...[truncated 56 chars]

Remediation
View remediation

Remediation Suggestions

  • Remove the irm | iex installation command.
  • Distribute the CLI through a reputable Windows package manager using a pinned version, or provide a versioned installer artifact.
  • Require Authenticode signature validation and verify that the signer certificate chains to the documented publisher.
  • If a standalone archive is used, verify a pinned cryptographic checksum obtained from an independent trusted channel before extraction or execution.
  • Avoid Invoke-Expression; save verified content locally and invoke only the specifically reviewed script or executable.
  • Run installation without administrative privileges unless a documented installation step strictly requires elevation.
  • Require informed user approval before installing software and document all expected system modifications.

SkillSpector

Vulnerability Patterns
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (2)

External Script Fetching

High
Category
Supply Chain
Confidence
95% confidence
Finding

The first-time setup section recommends piping a remotely fetched script directly into bash, which creates a supply-chain and remote code execution risk if the install endpoint, transport, or publishing pipeline is compromised. In a skill context, this is more dangerous because the content is operational guidance to be followed by an agent or user during troubleshooting, not merely documentation.

Content

Scanner excerpt · SKILL.md (reported line 57)May include surrounding context.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The description instructs the agent to use this skill for ANY MailboxValidator request and instead of calling the API directly, which is an overly broad trigger phrase. This can cause unintended invocation in contexts where the user only mentions MailboxValidator incidentally, increasing the chance of unnecessary tool use or surprising behavior.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.