Stormglass

• Outcome: Review
• Latest audit: Nov 4, 2026 · 5:40 AM UTC
• Version: 1.0.4

Overview

The Stormglass read workflow fits its purpose, but its setup instructions can trigger unverified remote installer execution without clear user approval.

Review setup before installing. Install a verified CLI release yourself or require approval before the agent runs an installer. Check the CLI's automatic skill-synchronization behavior and connect only the intended Stormglass account. The artifact contains no evidence of deliberate deception or exfiltration, but the remote installer contents and resulting system changes remain unverified.

A.I.G

Vulnerability Patterns

• Skill Instruction Hijacking: Alters the agent's session goals or safety constraints when the skill loads
• Agent Memory Poisoning: Writes attacker-controlled rules into memory that affect later sessions
• Remote Payload Retrieval and Execution: Fetches external code whose behavior can change after review
• Embedded Malicious Code: Ships malicious scripts inside the skill and executes them locally
• Unauthorized Access and Privilege Escalation: Obtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns

• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code
• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep
• Output Handling: Unvalidated Output Injection, Cross-Context Output, Unbounded Output
• MCP Tool Poisoning: Hidden Instructions, Unicode Deception, Parameter Description Injection
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands

Findings (5)

Description-Behavior Mismatch

High

• Category: MCP Tool Poisoning
• Confidence: 99% confidence

Finding

Installing the connector CLI is relevant to the skill, but the fallback expands a read-only data task into downloading and executing unverified remote code without explicit installation approval. If the installation endpoint or its delivery infrastructure is compromised, that code can execute with the agent process's privileges; the missing-command condition limits exposure but does not remove it.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High

• Category: MCP Tool Poisoning
• Confidence: 84% confidence

Finding

The supplied code performs only a login operation and shows no Stormglass integration or data-reading behavior. Authentication could be a supporting implementation detail, but nothing supplied connects this CLI login to Stormglass or demonstrates the declared functionality.

Content

No source excerpt is available for this finding.

External Script Fetching

High

• Category: Supply Chain
• Confidence: 98% confidence

Finding

Piping a mutable HTTPS response directly into bash executes code without version pinning, signature verification, or inspection. HTTPS protects transport but does not prevent malicious changes at the source, so compromise of the installer distribution path could lead to arbitrary code execution under the agent's account.

Content

Scanner excerpt · SKILL.md (reported line 57)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

The setup workflow executes downloaded installation scripts without warning the user about the system-level effects.

Medium

• Category: Output Handling
• Confidence: 96% confidence

Finding

The setup instructions direct the agent to execute downloaded shell or PowerShell code without explaining its installation effects or requiring user consent. A request to read weather or tide data does not itself authorize arbitrary installer execution, which can modify local files and access resources available to the executing user.

Content

No source excerpt is available for this finding.

Autonomous Decision Making

Medium

• Category: Excessive Agency
• Confidence: 80% confidence

Finding

Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Content

Scanner excerpt · SKILL.md (reported line 78)May include surrounding context.

md

- **`OOMOL_INSUFFICIENT_CREDIT`** — OOMOL billing stop. Recharge at `https://console.oomol.com/billing/token-recharge` before retrying.

- **HTTP 402 without `OOMOL_INSUFFICIENT_CREDIT`** — the third-party account itself has no active plan or credit. Report the provider error message as is; do not ask the user to recharge OOMOL.

## Resources