skills/letta-ai/skills/ai-news/Gen Agent Trust Hub

ai-news

Pass

Audited by Gen Agent Trust Hub on Apr 23, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill fetches AI news updates from reputable industry RSS feeds (OpenAI, Google, Hugging Face, etc.) and uses YouTube's public RSS interface to track channel updates.\n- [COMMAND_EXECUTION]: Executes purpose-built TypeScript scripts (fetch-news.ts and fetch-transcript.ts) using npx tsx to retrieve and format external news data.\n- [PROMPT_INJECTION]: The skill processes external news items and transcripts (Indirect Prompt Injection surface). Ingestion points: fetch-news.ts and fetch-transcript.ts. Boundary markers: Results are returned as structured JSON to the agent. Capability inventory: Network access for data fetching and script execution. Sanitization: The stripHtml function removes HTML tags from fetched content summaries.\n- [SAFE]: The skill's implementation is transparent and consistent with its described purpose. No sensitive file access, credential exposure, or obfuscated code patterns were found.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 23, 2026, 06:38 AM