Back to skill

Security audit

Gmail

Security checks for vulnerabilities and agentic risk

Overview

This Gmail skill needs review because setup executes downloaded code immediately and attachment uploads bypass its confirmation safeguards.

Review and approve CLI installation separately, preferably using a verified release. Require confirmation before attachments enter transit storage and check its access and retention rules. Review the CLI's automatic skill synchronization before first use. The remote installers and hosted connector implementation were not verified; no artifact-backed malicious behavior was found.

Vulnerability Patterns
  • 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
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Vulnerability Patterns
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Findings (4)

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
97% confidence
Finding

The supplied chunk performs remote script execution rather than any demonstrated Gmail operation. Although installing a CLI can be a supporting implementation detail, nothing supplied establishes this installer as a Gmail dependency. The remote script's contents are unavailable, so its further behavior and any Gmail functionality cannot be verified.

Content

No source excerpt is available for this finding.

External Script Fetching

High
Category
Supply Chain
Confidence
98% confidence
Finding

The setup command executes a remotely fetched, unpinned shell script without inspection or independent integrity verification; the PowerShell alternative has the same weakness. Although installation is a legitimate prerequisite and only occurs when the CLI is missing, a compromised distribution endpoint could execute arbitrary code with the agent's local permissions and expose accessible files or credentials.

Content

Scanner excerpt · SKILL.md (reported line 101)May include surrounding context.

The untagged attachment action uploads data despite the safety claim that all untagged actions are reads.

Medium
Category
Excessive Agency
Confidence
97% confidence
Finding

The untagged download_attachment action uploads attachment contents to Connector file transit storage, yet the safety section categorizes all untagged actions as reads safe to run directly. This can cause sensitive email attachments to be copied into additional storage without confirmation of that transfer; the existing use of the same connector provider reduces evidence of malicious exfiltration but does not eliminate the privacy risk.

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 122)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