seo-content-brief

Pass

Audited by Gen Agent Trust Hub on Aug 6, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes the belt CLI tool to perform search, content extraction, and image generation. It implements a restricted execution environment via the frontmatter configuration allowed-tools: Bash(belt *), which limits the agent to only running commands prefixed with belt, reducing the risk of arbitrary command injection.
  • [EXTERNAL_DOWNLOADS]: The skill provides instructions for installing external tools including the belt-sh/cli package and several related skills from the inference-sh organization using npx. It also references installation documentation hosted on GitHub at https://raw.githubusercontent.com/inference-sh/skills/refs/heads/main/cli-install.md.
  • [PROMPT_INJECTION]: The skill involves an indirect prompt injection surface as it is designed to ingest and summarize data from external web sources. Ingestion points: Search results and web content retrieved via the tavily/search-assistant, exa/search, and tavily/extract apps. Boundary markers: None identified in the provided prompt templates to isolate retrieved content from instructions. Capability inventory: Access to restricted Bash commands and external inference.sh applications. Sanitization: No explicit sanitization or filtering of external content is described in the instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 6, 2026, 05:03 AM
Security Audit — agent-trust-hub — seo-content-brief