setup-architect
Warn
Audited by Gen Agent Trust Hub on Feb 21, 2026
Risk Level: MEDIUMEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- EXTERNAL_DOWNLOADS (MEDIUM): The skill directs users to clone a repository from an untrusted GitHub account (codenamev/ai-software-architect) to provide the framework files and installation instructions.
- COMMAND_EXECUTION (MEDIUM): The skill uses the Bash tool to automate installation and cleanup tasks, including the removal of .git directories; such operations are high-risk when performed on content derived from untrusted external sources.
- PROMPT_INJECTION (LOW): The skill analyzes local project code to generate its 'Initial System Analysis' report, creating an indirect prompt injection vulnerability where malicious instructions embedded in the analyzed code could influence the agent's architectural recommendations. Evidence: 1. Ingestion points: project files accessed via Read/Glob/Grep. 2. Boundary markers: Absent. 3. Capability inventory: Bash, Write, Edit tools. 4. Sanitization: Absent.
Audit Metadata