architecture-explorer
Pass
Audited by Gen Agent Trust Hub on Oct 16, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill uses specific CodeGraph tools (
codegraph_files,codegraph_explore,codegraph_callers, etc.) to analyze architectural patterns. These operations are limited to scanning the project's own structure and metadata. - [SAFE]: Reports generated by the skill are written to standard temporary directories (
/tmpor$TMPDIR), ensuring that no persistent or unauthorized file modifications are made to the repository. - [INDIRECT_PROMPT_INJECTION]: The skill processes source code and existing documentation to infer module responsibilities. While processing untrusted codebase content can pose a theoretical surface for indirect prompt injection, the skill's instructions strictly focus the agent on structural analysis and architectural health assessments, minimizing the potential for malicious instruction execution.
- [SAFE]: The skill references other related skills (
project-wiki,codebase-design) to provide a consistent vocabulary and workflow, which is a standard pattern for multi-skill agent environments.
Audit Metadata