Security audit

Cron Setup

Security checks for vulnerabilities and agentic risk

Overview

This skill is a cron-job helper, but it tells scheduled jobs to send outputs to a fixed Telegram destination, which could repeatedly expose user data outside the local environment.

Review any generated cron job before enabling it, especially the delivery section and task message. Replace the hard-coded Telegram target with a destination you control, avoid sending secrets or private workspace data, and audit or delete any recurring jobs created from this skill if the recipient was not explicitly intended.

A.I.G

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

Findings (1)

T01 · Skill Instruction Hijacking

Error

Location: SKILL.md:64
Finding: Hard-Coded Telegram Destination Redirects Scheduled Task Output to an External Recipient
Content: View full analysis
Remediation: View remediation

SkillSpector

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access

Findings (3)

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: The description says to use the skill for 'periodic tasks, reminders, automated checks, or any scheduled work,' which is broad and overlaps with many common scheduling requests. It does not provide narrower trigger phrases, exclusions, or negative examples to clarify when this skill should be selected versus other automation skills.
Content: No source excerpt is available for this finding.

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 93% confidence
Finding: The skill explicitly instructs cron jobs to send messages to a specific Telegram target and thread without any user-facing consent, data-classification guidance, or warning that job outputs may leave the local environment. In a cron/scheduled context, this is more dangerous because autonomous tasks may repeatedly exfiltrate workspace contents, alerts, or command results to an external channel without an interactive checkpoint.
Content: No source excerpt is available for this finding.

Intent-Code Divergence

Low

Category: Not specified by scanner
Confidence: 92% confidence
Finding: The documentation states the default model for cron jobs should be anthropic/claude-sonnet-4-5 and explicitly says not to use DeepSeek for cron jobs, but the canonical JSON job template uses openrouter/deepseek/deepseek-v3.2 as its model. This is an active contradiction in the skill's own guidance and could cause users to configure jobs contrary to the stated convention.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.