Security audit

Dissertation Workflow Core

Security checks for vulnerabilities and agentic risk

Overview

This is a dissertation-writing workflow skill whose private-data and Zotero use are disclosed and aligned with its academic support purpose.

Install only if you intend the agent to work with Ryan Tang's dissertation materials, including potentially sensitive interview data, completed chapters, mentor feedback, Google Drive project files, and Zotero references. Confirm before allowing Zotero additions or broader searches, and treat the English/style constraints as workflow defaults you may override in your own prompts.

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

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
Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep

Findings (2)

Natural-Language Policy Violations

Medium

Confidence: 95% confidence
Finding: The TRANSLATE workflow explicitly requires translating Chinese text into academic English, and the broader skill description centers dissertation writing support in English. This imposes a specific language output without any opt-in or alternative language choice, which matches the policy concern for language/locale restrictions in natural-language instructions.

Natural-Language Policy Violations

Low

Confidence: 82% confidence
Finding: The instruction to draft specifically in Lois Weis's academic style constrains output style in a way that may override user preference. While the dissertation context may partially justify it, the file does not present this as a user-selectable preference or clearly document it as an optional default.

Static analysis

No suspicious patterns detected.