improve-codebase-architecture
Pass
Audited by Gen Agent Trust Hub on Apr 12, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [SAFE]: The primary function of the skill is codebase analysis and architectural guidance, which are standard, benign operations for a development agent.\n- [COMMAND_EXECUTION]: The skill integrates with the GitHub CLI (
gh) to document RFCs as issues. This action is gated by user input and occurs only after a design candidate has been selected and refined through multiple steps of analysis.\n- [PROMPT_INJECTION]: The skill is susceptible to Indirect Prompt Injection because it analyzes untrusted codebase files. This risk is inherent to tools that process external code but is mitigated here by a human-in-the-loop workflow requiring explicit user approval at checkpoints. - Ingestion points: Broad codebase exploration as described in SKILL.md.
- Boundary markers: None explicitly defined in the instructions.
- Capability inventory: Utilizes
gh issue createand local file writing to the./plans/directory. - Sanitization: No specific content sanitization or instruction filtering is applied to the codebase data.
Audit Metadata