Back to skill

Security audit

Magnific (Freepik)

Security checks for vulnerabilities and agentic risk

Overview

The skill mostly matches its Freepik connector purpose, but its setup instructions include unverified internet-downloaded installer scripts that could run arbitrary code.

Review the setup path before installing. Prefer an official, verifiable oo CLI install method, avoid running the installer in an elevated shell, and understand that using the skill routes requests through an OOMOL-connected Freepik account and may rely on account scopes, credentials, billing, or credits.

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:57
Finding
Unverified Remote Installer Download and Execution## Vulnerability Details **File Location**: `SKILL.md`, lines 57–61 **Vulnerability Type**: 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 `cli.oomol.com` and immediately execute them through Bash or PowerShell. Neither command pins a specific installer version, validates a cryptographic checksum or signature, nor gives the user an opportunity to inspect the downloaded code before execution. Although installing the `oo` CLI is related to the Skill's declared functionality, direct pipe-to-shell execution is not the minimum privilege or safest mechanism necessary to perform that installation. The effective code is hosted outside the audited project and may change at any time after the Skill has been reviewed. The Skill does not itself demonstrate that the current remote scripts are malicious. However, compromise of the hosting service, publication pipeline, DNS or TLS trust path, or vendor account would turn these documented commands into an arbitrary-code-execution channel. ### Attack Path 1. The `oo` command is unavailable and an agent or user follows the first-time setup instructions. 2. The shell requests the current `install.sh` or `install.ps1` response from `cli.oomol.com`. 3. The response is passed directly to Bash or PowerShell without integrity verification or local inspection. 4. A compromised or unexpectedly modified remote response executes with the privileges of the invoking process. 5. The executed payload can access files and credentials available to that account, modify the environment, install additional components, or establish persistence. ### Impact Assessment Successful exploitation provides arbitrary command executio ...[truncated 607 chars]
Remediation
## Remediation Suggestions 1. Remove the direct `curl | bash` and `irm | iex` installation patterns. 2. Direct users to a pinned, versioned release hosted in the vendor's official release repository. 3. Download the installer or package to a local file without executing it automatically. 4. Publish and verify a cryptographic checksum and, preferably, a platform-native digital signature before execution. 5. Display the resolved version, source URL, checksum, and intended installation changes for user review. 6. Require explicit user approval before executing any installer. 7. Run installation with ordinary user privileges unless a specific, documented operation requires elevation. 8. Prefer signed platform package managers or reproducible package artifacts where available. 9. Fail closed if signature or checksum verification cannot be completed; do not silently fall back to remote shell 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
97% confidence
Finding
The skill instructs the agent to install software by piping a remotely fetched script directly into bash, which creates a classic supply-chain and remote code execution risk. If the install endpoint, transport, or hosting environment is compromised, arbitrary code would run immediately on the user's system without verification.

Vague Triggers

Medium
Confidence
97% confidence
Finding
The instruction to use this skill for ANY Magnific (Freepik) request is overly broad and can cause the agent to invoke the skill on simple mentions or weakly related tasks without sufficient user intent verification. In context, this increases the chance of unintended connector usage and unnecessary exposure to external service calls, even though the listed actions are currently read-oriented.

Static analysis

No suspicious patterns detected.