skills/lev-os/agents/research/Gen Agent Trust Hub

research

Pass

Audited by Gen Agent Trust Hub on May 11, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: Spawns various search and extraction tools (brave-search, firecrawl, valyu) and local helper scripts to automate the research workflow.
  • [EXTERNAL_DOWNLOADS]: Fetches data from multiple web, news, and academic sources via established API endpoints (OpenAI, xAI, Perplexity, arXiv, GitHub).
  • [PROMPT_INJECTION] (Indirect): Processes large amounts of untrusted content from the web and social media. The skill mitigates risks associated with this ingestion surface by using structured synthesis templates and evidence-based reporting.
  • [CREDENTIALS_UNSAFE] (Guidance): Provides clear instructions for users to manage API keys securely in local .env and configuration files (e.g., ~/.env.local).
Audit Metadata
Risk Level
SAFE
Analyzed
May 11, 2026, 04:18 PM
Security Audit — agent-trust-hub — research