visual-feedback

Pass

Audited by Gen Agent Trust Hub on Jun 16, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill requires downloading and installing the agentation package from the npm registry as a development dependency.
  • [COMMAND_EXECUTION]: The skill executes shell commands to perform framework detection and install necessary dependencies using package managers (npm).
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by processing external data to drive code modifications.
  • Ingestion points: Annotations including element selectors and user comments are read from a local browser session file at runtime (as described in SKILL.md).
  • Boundary markers: The instructions do not specify the use of delimiters or boundary markers to isolate the user-provided annotation data from the agent's instructions.
  • Capability inventory: The skill allows the agent to install packages, detect project frameworks, and modify local source code files based on external input.
  • Sanitization: There is no mention of sanitization or validation of the user comments before they are used to guide the surgical code edits.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 16, 2026, 10:10 PM
Security Audit — agent-trust-hub — visual-feedback