skills/hsiangjenli/skills/tech-slide/Gen Agent Trust Hub

tech-slide

Pass

Audited by Gen Agent Trust Hub on Oct 9, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The file scripts/check_dependencies.py includes a print statement suggesting the installation of the uv tool using a piped shell command (curl -LsSf https://astral.sh/uv/install.sh | sh). This command targets a well-known service (Astral) and is provided as a user instruction rather than being executed automatically by the skill.
  • [COMMAND_EXECUTION]: The script scripts/check_dependencies.py uses subprocess.run to call the uv CLI tool. These calls are used for routine environment synchronization (uv sync) and version checking, which are standard operations for managing project dependencies.
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to process user-provided technical content to generate presentation slides, which introduces a surface for indirect prompt injection.
  • Ingestion points: The agent modifies and processes slides.md based on user-supplied information.
  • Boundary markers: The skill utilizes standard Markdown frontmatter and slide delimiters (---) to structure content.
  • Capability inventory: The skill performs file copying (cp), dependency syncing (uv sync), and slide rendering (npx slidev).
  • Sanitization: No specific input sanitization is implemented; the skill relies on the Slidev parser to handle the markdown content.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 9, 2026, 08:10 AM