Security audit

Boss Battle Calendar

Security checks for vulnerabilities and agentic risk

Overview

This skill is a low-impact planning/template skill with no evidence of hidden access, persistence, exfiltration, or destructive behavior.

Installers should expect a lightweight text-planning helper, not an automated calendar/reminder system. It may produce guidance rather than fully computed schedules, but it does not show signs of accessing private data or changing external systems.

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

MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
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
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access

Findings (5)

Tp4

High

Category: MCP Tool Poisoning
Confidence: 98% confidence
Finding: There is a clear description-behavior mismatch. The declared purpose describes a user-facing planning/transformation skill for high-stakes events. However, the code is a generic markdown-template renderer: it loads SKILL.md, extracts headings/sections, normalizes input, and formats a response listing sections such as Purpose, Use cases, Workflow, and Output format. The only substantive action is local file reading of the skill definition. While it states it does not perform external actions, the main issue is that its primary behavior is documentation/template summarization, not the specialized event-preparation assistant described.
Content: No source excerpt is available for this finding.

Tp4

High

Category: MCP Tool Poisoning
Confidence: 95% confidence
Finding: The declared description says the skill should help users prepare for exams, trips, deadlines, meetings, or other date-driven events by converting them into boss-battle style plans. However, the supplied code chunk does not implement that behavior. It is a test file that checks for presence of metadata, prompt-template headings, and that the handler returns a descriptive card referencing SKILL.md and the user's input. Based on this code, the observable functionality is around loading and presenting skill documentation/template content, not producing the specialized countdown/checkpoint/risk/recovery planning described. This is a material purpose mismatch, not just an implementation detail.
Content: No source excerpt is available for this finding.

Undeclared Tool Scope

Medium

Category: MCP Least Privilege
Confidence: 70% confidence
Finding: Without declared permissions the skill's intent is opaque and cannot be validated.
Content: No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium

Category: Not specified by scanner
Confidence: 97% confidence
Finding: The manifest describes a skill that transforms upcoming high-stakes events into boss-battle style plans with countdown phases, checkpoints, warnings, and recovery plans. In practice, the handler does not analyze dates or generate event-specific battle plans; it loads SKILL.md, extracts headings, echoes the user's input, and formats a static descriptive card from template sections.
Content: No source excerpt is available for this finding.

Intent-Code Divergence

Low

Category: Not specified by scanner
Confidence: 80% confidence
Finding: The status text says the handler 'does not perform external actions,' which implies a purely self-contained response. However, the code reads SKILL.md from the filesystem via _read_skill_file before building the result, so the documentation overstates the absence of side effects or resource access.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.