skills/szoloth/skills/agentic-review/Gen Agent Trust Hub

agentic-review

Pass

Audited by Gen Agent Trust Hub on May 5, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [SAFE]: No malicious patterns, exfiltration, or obfuscation were detected. The skill uses standard local utilities for its stated purpose.
  • [COMMAND_EXECUTION]: Executes git commands like git diff and git status to identify local changes for analysis.
  • [PROMPT_INJECTION]: As a tool that processes raw code changes, it is vulnerable to indirect prompt injection where instructions hidden in code comments could influence analysis sub-agents.
  • Ingestion points: Raw code diffs from local git repositories.
  • Boundary markers: None identified in the prompt templates sent to sub-agents.
  • Capability inventory: Shell command execution (git) and file-write capabilities for applying fixes.
  • Sanitization: No filtering or sanitization of input code content is performed.
Audit Metadata
Risk Level
SAFE
Analyzed
May 5, 2026, 04:02 PM
Security Audit — agent-trust-hub — agentic-review