Back to skill

Security audit

Ask The Way 问道

Security checks for vulnerabilities and agentic risk

Overview

This is a low-risk reflective Dao De Jing guidance skill with no code execution or data access, but users should not treat it as therapy, crisis help, or professional advice.

Install only with the expectation that this is a philosophical reflection tool. Do not use it for emergencies, crisis support, therapy, diagnosis, legal, medical, financial, factual, or time-sensitive decisions; ask for a single-language response if bilingual output is not wanted.

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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The README explicitly positions the skill for 'life decisions,' 'existential uncertainty,' and 'anxiety' but does not warn that outputs are philosophical reflection rather than professional mental-health, medical, or crisis guidance. That can cause vulnerable users to over-rely on the skill, delay appropriate support, or treat poetic advice as clinically suitable help.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The checklist requires bilingual output as a fixed behavior, and the metadata also presents the skill as Chinese+English. This can violate language/locale policy when users have not opted into bilingual responses or been given a language preference choice.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.