deep-review

Pass

Audited by Gen Agent Trust Hub on May 9, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill provides instructions to download missing dependencies from the author's repository (github.com/mikeng-io/agent-skills) using git clone if they are not present locally.\n- [COMMAND_EXECUTION]: Utilizes Bash for repository management (git), directory navigation (ls), and creating output folders (mkdir). It also references the qmd CLI for optional artifact management.\n- [PROMPT_INJECTION]: The skill processes user-provided content and artifacts which are interpolated into reviewer agent prompts, creating an indirect prompt injection surface.\n
  • Ingestion points: User artifacts and conversation history identified in the working_scope.\n
  • Boundary markers: Prompts use markdown headers (e.g., ## Context to Review) to isolate user content.\n
  • Capability inventory: Capabilities include writing files (Write), shell execution (Bash), and sub-task spawning (Task).\n
  • Sanitization: No explicit sanitization or escaping of interpolated content is implemented beyond standard agent guardrails.
Audit Metadata
Risk Level
SAFE
Analyzed
May 9, 2026, 05:42 PM