Security audit

ParseHub

Security checks for vulnerabilities and agentic risk

Overview

The skill is mostly coherent for read-only ParseHub access, but it asks users to install a third-party CLI with unverified pipe-to-shell commands and broadly routes ParseHub activity through OOMOL.

Review before installing. Normal ParseHub reads are limited to get/list actions through the oo CLI, but you should only proceed if you are comfortable routing ParseHub request and response data through OOMOL. Do not run the listed curl|bash or irm|iex installer commands unless you independently trust and verify the OOMOL CLI installer source; prefer a signed or package-manager install path if available.

A.I.G

Vulnerability Patterns

Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
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:54
Finding: Unverified Remote Installer Download and Execution
Content: View full analysis
): ```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 installation instructions download mutable content from an external server and pass it directly to a command interpreter. Neither command pins an installer version, validates a cryptographic checksum or signature, nor gives the user an opportunity to inspect the downloaded script before execution. Although installing the `oo` CLI supports the declared functionality, immediate pipe-to-shell execution is not the minimum privilege or minimum-risk installation method. The effective code is controlled by the remote server and can change after this Skill has been reviewed. Compromise of the download host, its deployment pipeline, DNS resolution, or the TLS endpoint could therefore turn these documented installation commands into an arbitrary-code execution channel. The instructions cover both macOS/Linux and Windows, so the exposure is cross-platform. The Skill states that setup should only occur after a matching command failure, which reduces invocation frequency but does not mitigate the integrity risk. ### Attack Path 1. A user attempts a ParseHub operation and receives an `oo: command not found` error. 2. The user or agent follows the first-time setup instructions. 3. `curl` or `Invoke-RestMethod` retrieves the current installer from `cli.oomol.com`. 4. The downloaded response is passed directly to Bash or PowerShell without integrity verification. 5. If the hosting or delivery path has been compromised, attacker-controlled commands execute ...[truncated 921 chars]
Remediation: View remediation
" "oo-install.sh" | sha256sum --check - less oo-install.sh bash oo-install.sh ``` The example endpoint and checksum must be replaced with an immutable official release and its independently published checksum. ]]>

T01 · Skill Instruction Hijacking

Warning

Location: SKILL.md:1
Finding: Unconditional Redirection of ParseHub Requests Through a Third-Party Connector
Content: View full analysis
" ``` **2. Run the action** with a JSON payload that matches the input schema: ```bash oo connector run "parsehub" --action "" --data '' --json ``` ``` ### Technical Analysis The Skill description directs the agent to use this Skill for “ANY” ParseHub request and explicitly forbids direct API use. This blanket instruction modifies the agent's handling of all ParseHub-related tasks rather than limiting itself to requests where the user has selected or consented to OOMOL as an intermediary. The documented execution model sends action names and JSON request payloads to an OOMOL-managed connector and returns ParseHub re ...[truncated 2124 chars]
Remediation: View remediation

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: 98% confidence
Finding: The skill instructs users to install software via a remote script piped directly into a shell (curl ... | bash), which executes unverified code from the network with no integrity check or review step. In a skill context, this is especially dangerous because the agent may surface or encourage a high-risk bootstrap action that can lead to arbitrary code execution if the host, transport, or distribution channel is compromised.
Content: Scanner excerpt · SKILL.md (reported line 56)May include surrounding context.

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

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 96% confidence
Finding: The description says to use this skill for "ANY ParseHub request" and "Whenever a task involves ParseHub," which is a very broad activation condition. It does not define scope boundaries or exclusions, so ordinary mentions of ParseHub could trigger the skill unintentionally.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.