stophy-download

Pass

Audited by Gen Agent Trust Hub on Nov 7, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill downloads content from arbitrary external websites and saves it as markdown files, creating a surface for indirect prompt injection if the agent later processes these files.
  • Ingestion points: External URLs provided to the stophy crawl command in SKILL.md examples.
  • Capability inventory: The skill uses Bash to create directories (mkdir) and write files (> "bun-docs/$name.md").
  • Boundary markers: No specific delimiters or instructions are provided to help the agent distinguish between its own instructions and the content downloaded from the web.
  • Sanitization: The skill performs filename sanitization using sed, but does not sanitize the markdown content itself.
  • [COMMAND_EXECUTION]: The skill executes shell commands to perform its primary functions, including site crawling and JSON processing.
  • Evidence: Core logic in SKILL.md uses stophy, jq, sed, and bash control structures.
  • Note: These commands are used as intended for the skill's purpose, and the allowed-tools metadata restricts the environment to the author's specific CLI tools.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 7, 2026, 09:19 PM