Back to skill

Security audit

Workday Music & Greet

Security checks for vulnerabilities and agentic risk

Overview

The skill's music-and-email purpose is mostly coherent, but it uses unvalidated shell commands and persistent cron setup around editable configuration and SMTP credentials.

Review before installing. This skill needs SMTP credentials, can send recurring emails, can switch local music scenes, and can create cron-based persistence. Only use it from a trusted, non-shared directory; inspect and keep scripts/config.json writable only by trusted users; update nodemailer; and replace the shell command construction with argument-array execution before enabling scheduled jobs.

Vulnerability Patterns
  • Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
  • 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
Findings (2)

T09 · Insecure Skill Coding Practices

Error
Location
scripts/scene-trigger.js:34
Finding

Shell Command Injection Through User-Editable Music Scene Configuration

Content
View full analysis
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Warning
Location
scripts/setup-cron.js:18
Finding

Command Injection Risk in Generated Cron Registration Commands

Content
View full analysis
Remediation
View remediation
23 || m < 0 || m > 59 ) { throw new Error(`Invalid scene time: ${scene.time}`); } ``` 3. Use a well-reviewed shell-escaping library for every dynamic argument if the output must be consumed by a shell. 4. Quote the Node.js executable and script path correctly inside the stored command. 5. Replace the secondary shell invocation in `scene-trigger.js` with argument-array execution: ```js const { execFileSync } = require('child_process'); execFileSync(process.execPath, [ path.join(__dirname, 'send-greet.js'), sceneKey, ], { stdio: 'inherit', timeout: 30000, shell: false, }); ``` 6. Display generated job definitions for explicit user review before registration. 7. Provide a documented command for listing and removing all jobs created by the Skill. 8. Register scheduled tasks under a least-privileged account. ]]>
Vulnerability Patterns
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • 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
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
Findings (12)

Tp4

High
Category
MCP Tool Poisoning
Confidence
91% confidence
Finding

The skill claims automatic scheduling, music control, and IMAP/SMTP functionality, but the described implementation appears inconsistent and may require manual command-line arguments instead of the promised workflow. This discrepancy is dangerous because deceptive or inaccurate capability claims can hide what the skill actually does, undermining informed consent and making malicious substitutions harder to detect.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

The skill claims automatic scheduling, music control, and IMAP/SMTP functionality, but the described implementation appears inconsistent and may require manual command-line arguments instead of the promised workflow. This discrepancy is dangerous because deceptive or inaccurate capability claims can hide what the skill actually does, undermining informed consent and making malicious substitutions harder to detect.

Content

No source excerpt is available for this finding.

Known Vulnerable Dependency: nodemailer==6.10.1 — 12 advisory(ies): CVE-2026-82661 (Nodemailer: CRLF injection in Nodemailer List-* header comments allows arbitrary); GHSA-2x7j-588g-ccc2 (Nodemailer: Quadratic (O(n²)) time complexity in addressparser allows remote den); GHSA-8m3c-c648-2xjj (Nodemailer: resolveContent() on a MailMessage bypasses disableFileAccess/disable) +9 more

High
Category
Supply Chain
Confidence
98% confidence
Finding

The lockfile pins nodemailer to 6.10.1, and the provided advisories indicate this version has multiple known security issues, including header injection, denial-of-service, and content access/bypass behaviors. In this skill's context, the package is directly relevant because the skill sends greeting emails, so vulnerable mail construction and parsing paths are more likely to be exercised and could enable message manipulation, abuse of outbound email behavior, or service disruption.

Content

No source excerpt is available for this finding.

Known Vulnerable Dependency: nodemailer==6.10.1 — 12 advisory(ies): CVE-2026-82661 (Nodemailer: CRLF injection in Nodemailer List-* header comments allows arbitrary); GHSA-2x7j-588g-ccc2 (Nodemailer: Quadratic (O(n²)) time complexity in addressparser allows remote den); GHSA-8m3c-c648-2xjj (Nodemailer: resolveContent() on a MailMessage bypasses disableFileAccess/disable) +9 more

High
Category
Supply Chain
Confidence
97% confidence
Finding

The dependency set permits resolution to nodemailer 6.10.1, which is reported as having multiple advisories including header injection, denial-of-service, and file-access bypass issues. This skill’s context makes the issue more dangerous because it explicitly sends greeting emails, so vulnerable mail construction and content handling paths are likely reachable in normal operation.

Content

No source excerpt is available for this finding.

Credential Access

High
Category
Privilege Escalation
Confidence
60% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · scripts/send-greet.js (reported line 7)May include surrounding context.

js
const path = require('path');
const fs = require('fs');
const nodemailer = require('nodemailer');
require('dotenv').config({ path: path.resolve(__dirname, '..', '.env') });

const config = JSON.parse(fs.readFileSync(path.join(__dirname, 'config.json'), 'utf8'));
const template = fs.readFileSync(path.join(__dirname, '..', 'assets', 'email-template.html'), 'utf8');

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
88% confidence
Finding

The skill advertises behavior that clearly requires access to environment-based secrets such as SMTP credentials, but it does not declare any explicit tool scope or permissions. This weakens reviewability and sandbox enforcement, making it easier for a skill to access sensitive configuration without users understanding the full trust boundary.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The trigger phrases are generic and action-oriented, which increases the chance of accidental or overly broad invocation in unrelated contexts. Because the skill can send emails and potentially execute local commands, unintended triggering could cause unwanted outbound messages or local side effects.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

This manifest file's description says the skill will 'Auto-switch music scenes by workday schedule with GIF greeting emails,' but it does not specify the activation conditions, scope, or limits of when the behavior runs. In a manifest context, this is vague enough to leave unclear what triggers the skill and when it should or should not act.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

This manifest/config file contains a natural-language locale setting, cronTimezone: "Asia/Shanghai", that forces a specific regional timezone for all scheduled scenes. The file does not indicate user opt-in, configurability, or a region-specific justification, which creates a locale-policy concern under the rule for natural-language policy violations.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

This script invokes shell commands with execSync using interpolated values from config and argv, which creates command-injection risk if scene names or related config entries are modified by an attacker. Although shelling out is functionally related to the stated workflow, using a shell is broader than necessary and materially increases risk because a compromised config.json or crafted scene value could lead to arbitrary local command execution.

Content

No source excerpt is available for this finding.

Unpinned Dependencies

Low
Category
Supply Chain
Confidence
74% confidence
Finding

Using a caret range for dotenv weakens build reproducibility and increases supply-chain risk because installs may pull different versions over time. There is no specific vulnerable dotenv version identified here, so the issue is hygiene-related rather than evidence of active compromise.

Content

Scanner excerpt · package.json (reported line 1)May include surrounding context.

json
{"name": "workday-music-greet", "version": "1.0.0", "description": "Auto-switch music scenes by workday schedule with GIF greeting emails", "private": true, "dependencies": {"nodemailer": "^6.9.0", "dotenv": "^16.3.0"}}

Unpinned Dependencies

Low
Category
Supply Chain
Confidence
74% confidence
Finding

Using a caret range for dotenv weakens build reproducibility and increases supply-chain risk because installs may pull different versions over time. There is no specific vulnerable dotenv version identified here, so the issue is hygiene-related rather than evidence of active compromise.

Content

Scanner excerpt · package.json (reported line 1)May include surrounding context.

json
{"name": "workday-music-greet", "version": "1.0.0", "description": "Auto-switch music scenes by workday schedule with GIF greeting emails", "private": true, "dependencies": {"nodemailer": "^6.9.0", "dotenv": "^16.3.0"}}