slidespeak-generator

Pass

Audited by Gen Agent Trust Hub on Mar 6, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill provides a surface for indirect prompt injection by processing external data obtained from web searches.
  • Ingestion points: The skill uses the web_search tool to collect information for presentation slides, as specified in the Content Expansion Strategy section of SKILL.md.
  • Boundary markers: There are no explicit delimiters or specific 'ignore' instructions implemented to prevent the agent from following malicious commands embedded in the search results.
  • Capability inventory: The skill utilizes the slidespeak_render tool, which communicates with the SlideSpeak API (api.slidespeak.co) and writes files to the local system in the ./outputs/ppt directory.
  • Sanitization: No content filtering or validation logic is applied to the untrusted web data before it is incorporated into the tool's configuration parameters.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 6, 2026, 04:20 PM