ts-review

Pass

Audited by Gen Agent Trust Hub on May 7, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits a vulnerability to indirect prompt injection by interpolating untrusted input ($ARGUMENTS) directly into the prompt context without boundary markers or sanitization.
  • Ingestion points: The variable $ARGUMENTS in SKILL.md is used to ingest user-provided code for review.
  • Boundary markers: None. The content is placed directly after the "Review:" label without delimiters (like triple backticks or XML tags) or instructions for the agent to treat the content as data only.
  • Capability inventory: No direct shell execution or network operations are defined in this skill, but the resulting agent behavior could be manipulated by instructions hidden in the code being reviewed.
  • Sanitization: No validation or filtering is applied to the input content.
Audit Metadata
Risk Level
SAFE
Analyzed
May 7, 2026, 11:21 AM
Security Audit — agent-trust-hub — ts-review