stage

Pass

Audited by Gen Agent Trust Hub on Sep 8, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill uses and references established tools and packages for slide generation, such as Marp, reveal.js, and Slidev, which are hosted on official repositories and package registries.
  • [COMMAND_EXECUTION]: The instructions provide commands for users to preview generated slides, such as npx @marp-team/marp-cli slide.md --preview. These are standard operations for the intended presentation workflow.
  • [INDIRECT_PROMPT_INJECTION]: The skill accepts user-provided content (e.g., project specifications, outlines) to generate slides, creating a potential surface for embedded instructions in the source data to influence the agent's behavior during the conversion process.
  • Ingestion points: Data enters through handoffs described in reference/handoffs.md (e.g., SCRIBE_TO_STAGE_HANDOFF).
  • Boundary markers: No specific delimiters or warnings for the agent to ignore instructions within user-provided outlines are defined.
  • Capability inventory: The skill produces executable code in Markdown and HTML formats for various slide frameworks as described in SKILL.md.
  • Sanitization: The skill does not explicitly describe sanitization or validation of the input content before processing.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 8, 2026, 06:57 AM
Security Audit — agent-trust-hub — stage