codex-slides-ai-presentation

Pass

Audited by Gen Agent Trust Hub on Jul 30, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill requires the installation of the codex-slides plugin from the nexu-io organization on the Codex Marketplace and references external documentation hosted on GitHub.
  • [REMOTE_CODE_EXECUTION]: The skill utilizes the @codex-slides/sdk Node.js package and a CLI tool (codex-slides) to programmatically generate and render presentations, which involves executing code locally to handle file exports and rendering logic.
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface due to its core functionality of processing external data to generate slide content.
  • Ingestion points: The skill reads local Markdown, JSON, YAML, Code, and PDF files, and performs multi-round web research via MCP tools to gather content for slide decks.
  • Boundary markers: There are no explicit instructions or delimiters provided to the agent to treat external content as untrusted or to ignore embedded instructions within the ingested data.
  • Capability inventory: The agent has the capability to write to the file system (PPTX/PDF exports), perform network research, and execute parallel rendering processes.
  • Sanitization: The skill lacks evidence of content sanitization, filtering, or validation before the ingested data is used to generate outlines and visual content.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 30, 2026, 01:38 AM
Security Audit — agent-trust-hub — codex-slides-ai-presentation