frontend-slides-frontend-slides
Pass
Audited by Gen Agent Trust Hub on Jul 3, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes local scripts such as scripts/extract-pptx.py, scripts/deploy.sh, and scripts/export-pdf.sh to perform core functionalities including PowerPoint extraction, Vercel deployment, and PDF generation.
- [EXTERNAL_DOWNLOADS]: The skill instructions prompt for the installation of external dependencies like python-pptx via pip and utilize npx to execute vercel and playwright, which involves downloading assets from well-known technology services.
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes external, untrusted content from PowerPoint files, existing HTML presentations, and images which can influence the agent's logic during the generation of new presentation files.
- Ingestion points: PowerPoint (.pptx) files processed by the extraction script in Phase 4, HTML files ingested for enhancement in Mode C, and images analyzed in Step 1.2.
- Boundary markers: There are no explicit instructions or delimiters used in SKILL.md to isolate processed external content from the agent's system instructions.
- Capability inventory: The skill has the ability to execute shell commands (bash), run python scripts, and perform network operations via the Vercel CLI.
- Sanitization: There is no evidence of input validation or sanitization for the data extracted from user-provided files before it is used in prompt interpolation.
Audit Metadata