scope-appropriate-architecture

Pass

Audited by Gen Agent Trust Hub on Apr 13, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [SAFE]: No malicious patterns or security threats were identified. The skill's operations, including codebase analysis and architectural recommendations, are consistent with its stated purpose.
  • [COMMAND_EXECUTION]: The skill instructs the agent to use local shell commands (find, wc, grep, and a python3 one-liner) to gather metrics from the codebase. These commands are static and used solely for measuring file counts, line counts, and dependencies.
  • [PROMPT_INJECTION]: The skill accesses and processes data from the user's codebase, such as README files and package configurations, to determine project tiers. This represents an indirect prompt injection surface. Ingestion points: codebase files via Read, Glob, and Grep tools. Boundary markers: none. Capability inventory: execution of local analysis commands. Sanitization: none.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 13, 2026, 08:32 AM