newsletter-curation

Pass

Audited by Gen Agent Trust Hub on Aug 31, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructs the user to install a CLI tool and additional skill dependencies from external repositories.
  • Evidence: Commands such as npx skills add belt-sh/cli and npx skills add inference-sh/skills@email-design are used to extend functionality.
  • Source: The resources are fetched from the inference-sh and belt-sh repositories on GitHub, which are associated with the platform provider.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from the open web through automated search tools, which could contain adversarial instructions.
  • Ingestion points: The skill utilizes search tools (tavily/search-assistant and exa/search) to pull content from the internet for the newsletter curation process (SKILL.md).
  • Boundary markers: Absent. The skill does not define delimiters or system-level constraints to isolate retrieved content from the agent's instructions.
  • Capability inventory: The skill has the ability to execute shell commands via the belt CLI, including posting to social media (x/post-create) and running further search applications (SKILL.md).
  • Sanitization: The instructions include a manual "Curation Quality Filter" checklist that guides the human curator to verify source credibility and relevance, serving as a human-in-the-loop review mechanism.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 31, 2026, 11:00 PM
Security Audit — agent-trust-hub — newsletter-curation