The Agent Skills Directory

[SAFE]: All identified behaviors in the skill files are consistent with legitimate software development and architectural design tasks. No malicious patterns such as obfuscation, exfiltration, or unauthorized persistence were found.- [COMMAND_EXECUTION]: The skill uses routine development commands including git checkout for branch management and grep for verifying that code imports align with architectural dependency rules. These commands are scoped to the local project environment.- [INDIRECT_PROMPT_INJECTION]: The skill ingests user-provided architectural documentation as the basis for code generation. While this represents a surface for indirect prompt injection, the skill implements a multi-phase workflow requiring human review of intermediate artifacts (the Module Map and Migration Manifest), which effectively mitigates the risk of automated malicious instruction execution.

architecture-scaffold