dotnet-graphify-dotnet

Warn

Audited by Gen Agent Trust Hub on Apr 11, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill downloads the graphify-dotnet global tool and clones the https://github.com/elbruno/graphify-dotnet.git repository.
  • [REMOTE_CODE_EXECUTION]: The skill executes code from the downloaded global tool and cloned repository through dotnet tool install and dotnet run.
  • [COMMAND_EXECUTION]: The skill performs various shell operations including tool installation, version checks, and running the graphify command on local directories.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it processes untrusted source code files from a repository to generate knowledge graphs. * Ingestion points: Local repository or subtree files accessed during the graphify run workflow. * Boundary markers: None explicitly defined to separate source code content from AI instructions during the semantic enrichment phase. * Capability inventory: Subprocess calls for dotnet tool install, git clone, and graphify run which executes the extraction pipeline. * Sanitization: No evidence of sanitization or filtering of the ingested source code before it is passed to AI providers for semantic enrichment.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Apr 11, 2026, 12:41 PM