Security audit

Daily Report

Security checks for vulnerabilities and agentic risk

Overview

This is a simple daily reporting skill with disclosed local logs and Telegram alerts, with privacy and timezone settings users should review.

Before installing, confirm you want reports on the listed Spain-time schedule, configure Telegram only if that channel is approved for your business data, and periodically review or delete the local memory logs if they contain operational details.

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

SkillSpector

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
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

Findings (5)

Missing User Warnings

Medium

Confidence: 96% confidence
Finding: The skill mandates immediate Telegram alerts for operational events without any privacy warning or constraints on what information may be transmitted externally. This can leak business activity, budget status, replies, or system errors to a third-party messaging platform, potentially exposing sensitive operational metadata.

Natural-Language Policy Violations

Low

Confidence: 95% confidence
Finding: The instruction to send the morning report at `9:30 AM Spain` imposes a specific locale/timezone in natural language. The file does not offer user opt-in or explain why Spain is required, which conflicts with the policy against forcing a specific language or locale without choice.

Natural-Language Policy Violations

Low

Confidence: 95% confidence
Finding: The line `Send at 9 PM Spain` specifies a fixed locale/timezone for report delivery. Because the skill does not present this as configurable or justify it as region-specific, it constitutes a natural-language locale policy issue.

Missing User Warnings

Low

Confidence: 90% confidence
Finding: The skill instructs saving daily logs to persistent memory files, which modifies user data/state without any explicit warning, confirmation, or retention guidance. While the stored content appears operational rather than highly sensitive, silent persistence can create privacy, audit, and accidental data retention risks.

Missing User Warnings

Low

Confidence: 88% confidence
Finding: The memory file format section directs the agent to save structured operational data to local markdown files without warning that this creates persistent storage. Even if the data is business-oriented, local files may expose workflow details, metrics, and notes to unintended readers or future processes.

Static analysis

No suspicious patterns detected.