diagramming-processes
Pass
Audited by Gen Agent Trust Hub on Sep 12, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process diagram source code and legacy codebase information. Because it also guides the agent to execute rendering commands, it creates a surface where malicious instructions embedded in a project's files could attempt to influence the agent's behavior during documentation.
- Ingestion points: Diagram source files and legacy codebases referenced in SKILL.md.
- Boundary markers: None explicitly defined.
- Capability inventory: Execution of repository-owned scripts, make targets, and build commands.
- Sanitization: No mention of validation or escaping for the diagram source content.
- [COMMAND_EXECUTION]: The instructions direct the agent to execute rendering commands found within a repository, such as make targets or specific scripts ('One repo command renders everything'). This capability involves running shell commands that are derived from the project being analyzed.
Audit Metadata