improve-codebase-architecture
Pass
Audited by Gen Agent Trust Hub on Oct 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses local shell commands to walk the commit history using
git logand opens the generated architectural report using system-appropriate commands such asxdg-open,open, orstart.\n- [EXTERNAL_DOWNLOADS]: The generated HTML report fetches resources from well-known technology services, including the Tailwind CSS CDN and the jsDelivr CDN for the Mermaid.js library.\n- [INDIRECT_PROMPT_INJECTION]: The skill identifies an attack surface for indirect prompt injection as it processes untrusted content from the codebase.\n - Ingestion points: Local source code files,
git logoutput,GLOSSARY.md, and Architecture Decision Records (ADRs) found in the repository.\n - Boundary markers: The instructions do not define specific delimiters or warnings to prevent the agent from following instructions embedded in the analyzed code.\n
- Capability inventory: The skill can write files to the operating system's temporary directory, execute shell commands to open those files, and spawn sub-agents for exploration.\n
- Sanitization: No explicit sanitization or escaping of the codebase content is specified before it is used in the report or processed by the agent.\n- [DYNAMIC_EXECUTION]: The skill generates an HTML file that executes JavaScript (Mermaid.js) at runtime to render diagrams. The Mermaid configuration is set to
securityLevel: "loose", which allows the rendering of certain HTML tags within diagrams. If malicious content from the codebase (such as file names or comments) is inserted into these diagrams without proper escaping, it could lead to the execution of arbitrary scripts in the context of the local file.
Audit Metadata