frontpage-vote

Pass

Audited by Gen Agent Trust Hub on Jun 16, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill provides instructions to install components via npx from the author's GitHub repository (DFectuoso/frontpage-sh-skills).
  • [COMMAND_EXECUTION]: Utilizes curl for fetching data and the mppx command-line tool for executing blockchain-based payments and API interactions.
  • [DATA_EXFILTRATION]: Communicates with https://www.frontpage.sh to transmit user interactions like votes, comments, and project ideas.
  • [PROMPT_INJECTION]: The skill ingests user-generated data (titles, bodies, and comments) from an external API, creating a surface for indirect prompt injection.
  • Ingestion points: Fetches data via GET /api/proposals and GET /api/proposals/{id}/comments.
  • Boundary markers: No specific delimiters or warnings for the agent are included for processing the ingested external strings.
  • Capability inventory: Possesses network access to the primary API domain via curl and mppx.
  • Sanitization: Content submissions are described as being moderated through OpenAI's omni-moderation service.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 16, 2026, 01:12 AM
Security Audit — agent-trust-hub — frontpage-vote