Back to skill

Security audit

life-gacha

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed local journaling and daily activity reminder tool, with purpose-aligned file writes and recurring tasks that users should knowingly approve.

Before installing, confirm you are comfortable with the skill storing daily notes locally under its data directory and creating recurring reminders or scheduled tasks. Review the card pool in config.yaml, and only approve scheduling if you want the daily and weekly automation.

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
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Findings (8)

Tp4

High
Category
MCP Tool Poisoning
Confidence
97% confidence
Finding

While the implementation-mismatch portion is likely a false positive, the finding correctly highlights undeclared file creation/modification of weekly records despite no declared permissions. Hidden or undocumented state changes make it harder for users and hosts to understand what the skill will persist locally, which can undermine trust and safe execution boundaries.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
85% confidence
Finding

While the implementation-mismatch portion is likely a false positive, the finding correctly highlights undeclared file creation/modification of weekly records despite no declared permissions. Hidden or undocumented state changes make it harder for users and hosts to understand what the skill will persist locally, which can undermine trust and safe execution boundaries.

Content

No source excerpt is available for this finding.

Anti-Refusal Statement

High
Category
Anti-Refusal
Confidence
70% confidence
Finding

Skill instructs the agent to never refuse or to always comply. Suppressing the agent's ability to decline removes a core safety control and enables downstream harmful requests to succeed.

Content

Scanner excerpt · SKILL.md (reported line 10)May include surrounding context.

md
This skill is invoked when the user types `/life-gacha`, or says anything related to drawing a card, evening check-in, or weekly review.

**Always respond in the same language the user is using.**

---

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
91% confidence
Finding

The skill instructs the agent to read and modify local files (config.yaml and data/.../week-NN.md) but does not declare any tool scope or permissions. That creates an authorization gap where file access is implied by prompt instructions rather than explicitly constrained, increasing the chance of unintended file reads/writes if the runtime grants broad defaults.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The invocation rule allows activation when the user says anything merely 'related to' drawing a card, evening check-in, or weekly review, which is overly broad. Broad natural-language triggers can cause accidental invocation and unintended execution of file-modifying or scheduled-task setup behavior without clear user intent.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

This code file contains natural-language strings that require a specific language/locale for use, including the module docstring and later CLI messaging. Under the policy, forcing a specific language without offering the user a choice or documenting a justified locale restriction is a violation.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The module docstring states the skill reads weekly records and generates an encouraging summary entirely in Chinese, and the output strings throughout the file are fixed to Chinese. This is a natural-language locale constraint with no opt-in, user choice, or documented region-specific justification.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
95% confidence
Finding

The module docstring and command-line messages are written entirely in Chinese, which imposes a specific language choice on users without offering an alternative or opt-in. The policy explicitly calls for flagging language or locale constraints when a skill forces a specific language without user choice.

Content

No source excerpt is available for this finding.