newsletter-curation

Pass

Audited by Gen Agent Trust Hub on May 5, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references installation instructions and related modular skills from the author's official GitHub repository (github.com/inference-sh/skills). These are verified vendor resources.
  • [COMMAND_EXECUTION]: Utilizes the belt CLI tool to run specific applications for web searching (Tavily, Exa), image generation, and social media management. The usage is constrained to the tool's intended functionality within the Bash(belt *) scope.
  • [PROMPT_INJECTION]: This skill presents an indirect prompt injection surface by design, as it fetches and processes content from external websites (news, blogs, forums).
  • Ingestion points: Data retrieved from the internet via tavily/search-assistant and exa/search tools.
  • Boundary markers: Absent; there are no specific delimiters or instructions provided to the agent to disregard instructions potentially embedded in the sourced content.
  • Capability inventory: The skill has access to the belt CLI for running further applications and performing social media posts (x/post-create).
  • Sanitization: Absent; the agent is expected to read and summarize the fetched content directly into the newsletter format.
Audit Metadata
Risk Level
SAFE
Analyzed
May 5, 2026, 03:19 AM
Security Audit — agent-trust-hub — newsletter-curation