ckm-slides

Pass

Audited by Gen Agent Trust Hub on Jun 24, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill references a Python script for searching slide components located at .claude/skills/design-system/scripts/search-slides.py. This script is an external dependency used to fetch copywriting formulas, layouts, and strategies from a separate design system skill directory.
  • [PROMPT_INJECTION]: The skill processes user input via the $ARGUMENTS variable in references/create.md, which is interpolated into a task block. This represents an indirect prompt injection surface.
  • Ingestion points: User input captured in $ARGUMENTS within references/create.md.
  • Boundary markers: The skill uses <task> XML-style tags as delimiters.
  • Capability inventory: The agent can perform command execution (Python) and generate HTML files.
  • Sanitization: No explicit sanitization or instructions to ignore instructions within the input are provided.
  • [EXTERNAL_DOWNLOADS]: The HTML template in references/html-template.md loads the Chart.js library from the well-known jsDelivr CDN. This is a common practice for including front-end dependencies in generated HTML presentations.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 24, 2026, 12:40 PM
Security Audit — agent-trust-hub — ckm-slides