arinhub-review-pr

Pass

Audited by Gen Agent Trust Hub on Feb 18, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • Prompt Injection (LOW): The skill is vulnerable to Indirect Prompt Injection (Category 8) as it processes untrusted external data from GitHub pull requests.
  • Ingestion points: The skill ingests PR code, descriptions, and linked issue content via its subagents (code-reviewer, octocode-research, etc.).
  • Boundary markers: No specific boundary markers or instructions to ignore embedded commands are defined when passing PR data to subagents.
  • Capability inventory: The skill can write to the local filesystem (~/.agents/logs/) and triggers subagents capable of submitting comments/reviews to GitHub.
  • Sanitization: There is no explicit sanitization or filtering of the PR content before it is processed by the LLM-based subagents.
  • Data Exposure & Exfiltration (SAFE): The skill writes to ~/.agents/logs/. While this involves local file access, it is restricted to a specific log directory and does not target sensitive system files or credentials.
  • External Downloads (SAFE): No external scripts or packages are downloaded at runtime. It relies on other skills assumed to be present in the agent's environment.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 18, 2026, 08:36 PM