skills/guyvdn/skills/reveal-md/Gen Agent Trust Hub

reveal-md

Pass

Audited by Gen Agent Trust Hub on Aug 27, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands using reveal-md.cmd to serve presentations locally and export them to PDF format.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes user-controlled Markdown data (slides.md). While this is the intended purpose, it represents a surface where embedded instructions could influence the agent during the slide creation or export process.
  • Ingestion points: Reads and processes slides.md for rendering and exporting.
  • Boundary markers: No explicit delimiters or warnings are provided to separate user-supplied Markdown content from agent instructions.
  • Capability inventory: The agent can execute shell commands (reveal-md.cmd) and read/write files.
  • Sanitization: No sanitization or validation logic is specified for the input Markdown files.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 27, 2026, 12:11 PM
Security Audit — agent-trust-hub — reveal-md