denuvo-slides
Warn
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: MEDIUMEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill clones an external repository from an unverified GitHub user
Eng0AI. - [REMOTE_CODE_EXECUTION]: The skill directs the agent to run installation and development commands (
pnpm install,pnpm dev) on content downloaded from an external source, which can trigger arbitrary code execution via repository scripts. - [COMMAND_EXECUTION]: Multiple commands involving network access and project deployment are executed, including
git,pnpm,vercel, andnetlify. - [INDIRECT_PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection due to lack of sanitization for ingested external content.
- Ingestion points: Clones external repository
Eng0AI/denuvo-slides-template.gitinSKILL.md. - Boundary markers: No delimiters or safety instructions are provided for handling the external data.
- Capability inventory: Significant capabilities including command execution and network operations across build and deploy steps.
- Sanitization: No validation or filtering is performed on the downloaded template files.
Audit Metadata