Back to skill

Security audit

HomeBox

Security checks for vulnerabilities and agentic risk

Overview

This HomeBox skill is mostly coherent, but its setup instructions tell an agent to run remote installer scripts directly through a shell, which users should review before installing.

Review the oo CLI installation yourself before running it, prefer a pinned or verified installer if available, and confirm any HomeBox write or delete action before allowing the skill to proceed.

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 (1)

T03 · Remote Payload Retrieval and Execution

Error
Location
SKILL.md:72
Finding
Unverified Remote Installer Execution Through Shell Pipelines## Vulnerability Details **File Location**: `SKILL.md`, lines 72–76 **Vulnerability Type**: `T03: Remote Payload Retrieval and Execution` **Risk Level**: Critical ```bash curl -fsSL https://cli.oomol.com/install.sh | bash # macOS / Linux ``` ```powershell irm https://cli.oomol.com/install.ps1 | iex # Windows PowerShell ``` ### Technical Analysis The first-time setup instructions download mutable scripts from external URLs and immediately execute their contents using `bash` or PowerShell's `Invoke-Expression`. No release version is pinned, and the instructions provide no checksum verification, cryptographic signature validation, or opportunity to inspect the downloaded files before execution. HTTPS protects the connection in transit when its trust assumptions hold, but it does not establish that the current server-side script is the same artifact that was reviewed. Compromise of the hosting service, publishing account, DNS infrastructure, TLS trust path, or installer build pipeline could therefore replace the effective payload after this Skill has been audited. Installing a required CLI can be legitimate, and these commands are only presented as a fallback when `oo` is unavailable. However, direct download-to-shell execution exceeds the minimum privilege and integrity requirements necessary to install that CLI safely. ### Attack Path 1. The `oo` command is unavailable, causing the Skill to use its first-time setup instructions. 2. An attacker compromises or gains control over the installer endpoint, its deployment pipeline, or another component capable of changing the returned script. 3. The agent retrieves the attacker-controlled response from `cli.oomol.com`. 4. The pipeline passes the response directly to `bash` or `Invoke-Expression` without integrity validation. 5. The malicious script executes with the permissions of the account running the agent. 6. The script can access or alter resources availa ...[truncated 943 chars]
Remediation
## Remediation Suggestions - Remove both direct download-to-shell pipelines. - Pin installation to a specific reviewed CLI version rather than a mutable installer endpoint. - Download the installer or package to a local file without executing it automatically. - Verify a vendor-provided cryptographic signature or a SHA-256 digest obtained through a separately trusted, immutable release channel. - Abort installation if verification fails; never silently continue with an unverified artifact. - Prefer signed packages distributed through the operating system's standard package manager or an authenticated release repository. - Display the resolved version, source URL, verification result, and intended installation changes before execution. - Require explicit user approval before running an installer. - Execute installation with ordinary user privileges where possible and request elevation only for a narrowly defined operation when strictly necessary. - Document a manual installation procedure so users can inspect the artifact before execution.
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
Content
- **`oo: command not found`** — install the oo CLI (other platforms: <https://cli.oomol.com/install-guide.md>):

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

  ```powershell
Confidence
98% confidence
Finding
The skill instructs the agent to install software by piping a remotely fetched script directly into a shell. This creates a supply-chain and remote-code-execution risk: if the remote endpoint, transport, or published installer is compromised, arbitrary code will run immediately on the host without review.

Vague Triggers

Medium
Confidence
95% confidence
Finding
The manifest description uses an extremely broad activation condition: any task that "involves HomeBox" should use this skill. For a markdown/manifest file, this lacks scope boundaries or negative examples and could overlap with many ordinary references to HomeBox rather than clearly defined actions.

Static analysis

No suspicious patterns detected.