thesis-defense-pptx
Warn
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: MEDIUMCOMMAND_EXECUTIONPRIVILEGE_ESCALATIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill's core functionality relies on executing a suite of local Python and PowerShell scripts. These scripts perform sensitive operations including reading local files, writing new documents, and controlling Microsoft PowerPoint via COM automation.
- [PRIVILEGE_ESCALATION]: The skill explicitly instructs the agent to run PowerShell scripts using the
-ExecutionPolicy Bypassflag. This command is used to circumvent the system's execution policy, which is a security feature designed to control the conditions under which scripts are loaded and run. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted content from user-provided thesis documents, which could contain malicious instructions designed to manipulate the agent's behavior.
- Ingestion points: Content is extracted from local PDF, LaTeX, BibTeX, and Markdown files using the
scripts/extract_thesis_context.pytool. - Boundary markers: There are no explicit instructions or delimiters provided to the agent to treat the extracted thesis content as untrusted data or to ignore any embedded instructions within that content.
- Capability inventory: The agent has extensive capabilities including shell command execution, local file system writes, and application automation via COM, which could be abused if an injection is successful.
- Sanitization: The extraction process does not include sanitization or filtering to detect or neutralize potential prompt injection patterns in the source documents.
Audit Metadata