Skills
skills/shipshitdev/skills/comment-mode/Gen Agent Trust Hub

comment-mode

Pass

Audited by Gen Agent Trust Hub on Apr 10, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill writes feedback to a local HTML file and opens it using the open command. This is a standard functionality for displaying generated previews to the user.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided drafts which acts as an ingestion point for untrusted data. 1. Ingestion points: User-provided text drafts described in the feedback workflow. 2. Boundary markers: Absent in the interpolation template. 3. Capability inventory: File-write to _private/views/ and open subprocess (SKILL.md). 4. Sanitization: The template's JavaScript uses textContent for rendering comments, providing protection against XSS in the comment text itself, though the primary draft content is interpolated into the HTML body.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 10, 2026, 09:45 AM