lcs-improve-architecture
Pass
Audited by Gen Agent Trust Hub on Aug 9, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill performs architectural analysis by reading local source files and writing reports to the
.lcs/directory. The behavior is consistent with the stated purpose of identifying code duplication and generating refactoring plans. - [DATA_EXPOSURE]: The skill scans the codebase to identify logical features and duplicated concerns. While this involves reading local files, the workflow includes explicit exclusions for sensitive directories such as
.git,node_modules, and build output folders. This behavior is necessary for the skill's primary function and is restricted to the local filesystem. - [INDIRECT_PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection as it ingests untrusted data from the project's source code files. Maliciously crafted comments or code patterns in the analyzed files could theoretically influence the agent's output. However, the structured multi-phase workflow (Scope Discovery, Feature Identification, Duplication Analysis) and the requirement for specific file/line citations serve as functional mitigations against accidental obedience to embedded instructions.
Audit Metadata