Back to skill

Security audit

Nyne.ai

Security checks for vulnerabilities and agentic risk

Overview

The skill is mostly a coherent Nyne.ai connector wrapper, but its fallback setup tells agents or users to run unverified remote installer scripts directly in a shell.

Review the setup instructions before installing. Prefer installing the oo CLI from verified official releases or a package manager with integrity checks, and do not let an agent run the curl|bash or irm|iex commands automatically. Normal Nyne.ai connector actions should still be checked for whether they read data or submit/write requests, especially enrichment and search submissions.

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:59
Finding
Unverified Remote Installer Scripts Executed Directly by Shells## Vulnerability Details **File Location**: `SKILL.md`, lines 59–68 **Vulnerability Type**: Remote payload retrieval and execution **Risk Level**: Critical ```markdown - **`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 irm https://cli.oomol.com/install.ps1 | iex # Windows PowerShell ``` ``` ### Technical Analysis The setup instructions retrieve mutable scripts from external URLs and immediately execute the returned content using `bash` or PowerShell `Invoke-Expression`. The instructions do not pin an installer version, verify a cryptographic checksum or signature, or provide an opportunity to inspect the downloaded script before execution. HTTPS protects data in transit but does not establish that the current script is the same payload that was reviewed. Compromise of the hosting infrastructure, domain, DNS configuration, TLS credentials, or vendor publishing account could replace the installer after publication. A server-side error or malicious response would also be treated as executable code. Installing the required CLI supports the Skill's declared functionality, but direct execution of an unverified network response is not the minimum privilege or safest installation mechanism. The downloaded payload receives the permissions of the user running the command. ### Attack Path 1. An attacker compromises the installer host, vendor publishing account, DNS configuration, or another component capable of controlling the response from `cli.oomol.com`. 2. The attacker modifies `install.sh` or `install.ps1` to include arbitrary malicious commands. 3. A user or agent encounters an `oo: command not found` error and follows the documented installation command. 4. `curl` or `Invoke-RestMethod` downloads the attacker-controlled response. 5. ...[truncated 1151 chars]
Remediation
## Remediation Suggestions - Remove the `curl | bash` and `Invoke-RestMethod | Invoke-Expression` installation patterns. - Prefer a trusted platform package manager or a version-pinned release artifact. - Publish cryptographic checksums and, preferably, signed release artifacts with documented signature verification. - Download the installer to a local file, verify its checksum or signature against an independently distributed trusted value, and only then execute it. - Pin the installer and CLI to an explicit reviewed version rather than retrieving a mutable latest installer. - Document the files, permissions, and system changes made by the installer. - Do not automatically install the CLI in response to a command failure. Require informed user approval before downloading or executing installation code. - Run installation with ordinary user privileges unless a narrowly documented operation specifically requires elevation. - For PowerShell, avoid `Invoke-Expression`; invoke a verified local script file using an appropriate execution policy. - Provide manual installation instructions so users can inspect the artifact and commands 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
96% confidence
Finding
The skill recommends installing software via a curl-pipe-to-shell command, which executes remote code immediately without integrity verification or review. If the remote host, transport, or installation script is compromised, an agent or user following this guidance could execute arbitrary code on the local system.

Vague Triggers

Medium
Confidence
95% confidence
Finding
The description says to use this skill for "ANY Nyne.ai request," including reading, creating, and updating data, whenever a task involves Nyne.ai. This is a very broad activation condition with no negative examples or constraints, which can cause unintended invocation for loosely related mentions of Nyne.ai.

Static analysis

No suspicious patterns detected.