Back to skill

Security audit

Gender-API.com

Security checks for vulnerabilities and agentic risk

Overview

The skill is coherent for Gender-API.com lookups, but it needs Review because its setup path executes unverified remote installer scripts and its lookups can process personal data without explicit confirmation.

Install only after reviewing the oo CLI installation method and preferring a signed, checksummed, or package-managed installer. Treat name, full-name, email, IP, locale, and caller-ID lookups as personal-data transfers to OOMOL and Gender-API.com, and confirm that the user intends that processing before running queries.

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 System Shells## Vulnerability Details **File Location**: `SKILL.md`, lines 59–63 **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 retrieve mutable scripts from `cli.oomol.com` and immediately execute the responses with Bash or PowerShell. Neither command pins a version, verifies a cryptographic checksum or signature, nor gives the user an opportunity to inspect the downloaded content before execution. HTTPS protects data in transit but does not guarantee that the current remote payload is the same payload that was reviewed. A compromised hosting service, publishing account, domain, certificate trust chain, or installer pipeline could therefore turn these installation commands into an arbitrary-code execution channel. Future changes made by the legitimate publisher would also execute without being covered by this audit. The behavior exceeds the minimum privileges needed to document or invoke the Gender-API connector because installing a CLI does not inherently require executing unverified, mutable network content directly in a shell. The project provides no local implementation or integrity metadata that constrains what these remote scripts can do. The Skill also directs the connector to transmit names, email addresses, and optional IP or caller-ID hints to an external service. That transfer is consistent with its declared Gender-API lookup functionality, and no evidence of covert exfiltration was found. Nevertheless, users should be informed before personal data is sent through OOMOL to Gender-API.com. ### Attack Path 1. The `oo` CLI is unavailable, causing the user or Agent to follow the first-time setup instructions. 2. An attacker compromises or gains control over the remote installer distributio ...[truncated 1205 chars]
Remediation
## Remediation Suggestions 1. Remove both direct pipe-to-shell installation commands. 2. Prefer a trusted operating-system package manager or an official repository with package-signing verification. 3. Pin the CLI to a specific reviewed version rather than retrieving an unversioned installer. 4. If direct artifact installation is unavoidable: - Download the installer or binary to a local file without executing it. - Obtain its expected SHA-256 checksum from a separately secured source. - Verify a publisher signature and certificate identity where supported. - Abort installation if any integrity or authenticity check fails. - Allow the user to inspect the downloaded script before execution. - Execute it only after explicit user approval. 5. Document the expected publisher, destination paths, permissions, and system changes. 6. Do not request elevated privileges unless a specific installation operation requires them; prefer a user-scoped installation. 7. Inform users before transmitting names, email addresses, IP addresses, caller identifiers, or other personal data, and identify both OOMOL and Gender-API.com as network recipients or processors.
Vulnerability Patterns
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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
Findings (3)

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 recommends installing software via a remote script piped directly into the shell (`curl ... | bash`), which executes unverified code from the network without integrity checking or review. If the distribution endpoint, transport, or upstream script is compromised, this becomes an immediate remote code execution path on the user's machine.

Vague Triggers

Medium
Confidence
94% confidence
Finding
The instruction to use this skill for ANY Gender-API.com request is overly broad and can cause the agent to activate the skill whenever Gender-API.com is merely mentioned, even if the user did not intend to invoke account-backed queries. In this context, that increases the chance of unnecessary third-party data disclosure, unintended billing/credit consumption, and privacy-impacting lookups against names or email addresses.

Natural-Language Policy Violations

Medium
Confidence
90% confidence
Finding
The skill presents gender inference from names and email addresses as a default capability without requiring explicit user confirmation, policy checks, or sensitivity gating. Because these inputs are personal data and the output is an inferred sensitive attribute, the skill can facilitate privacy-invasive or discriminatory processing, especially when run automatically from minimal prompts.

Static analysis

No suspicious patterns detected.