improve-codebase-architecture
Pass
Audited by Gen Agent Trust Hub on May 14, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill defines a clear, domain-specific vocabulary and process for improving codebase architecture. It relies on local files and user interaction to guide refactoring decisions.
- [INDIRECT_PROMPT_INJECTION]: The skill possesses a vulnerability surface for indirect prompt injection as it processes untrusted codebase data to perform its primary function.
- Ingestion points: Reads project source code, Architecture Decision Records (ADRs) in
docs/adr/, and domain documentation. - Boundary markers: No explicit delimiters or instructions are used to ignore embedded commands within the processed files.
- Capability inventory: The agent can update project documentation (
CONTEXT.md) and orchestrate sub-agents for design and exploration tasks. - Sanitization: No sanitization or filtering of the ingested codebase content was identified.
Audit Metadata