improve-codebase-architecture
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted content from the local codebase and domain documentation to generate architectural refactoring proposals.
- Ingestion points: The agent reads project files, the domain glossary (CONTEXT.md), and Architecture Decision Records (docs/adr/) during exploration.
- Boundary markers: The skill requires the agent to use a specific architectural glossary (LANGUAGE.md) and project-specific domain language (CONTEXT.md) for all suggestions, providing structural constraints.
- Capability inventory: The skill uses the Agent tool to spawn sub-agents for exploration and interface design, and it has the capability to create or update CONTEXT.md and ADR files based on its analysis.
- Sanitization: There are no explicit instructions for sanitizing or escaping the content read from the codebase before it is used in the agent's prompts or written to documentation files.
Audit Metadata