rhwp-advanced

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructs the agent or user to execute @nomadamas/k-skill via npx to retrieve full instructions and file listings. This is a vendor-owned utility used for instruction delivery.
  • [EXTERNAL_DOWNLOADS]: The skill requires the installation of the rhwp Rust CLI, suggesting either cargo install rhwp or downloading pre-built binaries from the edwardkim/rhwp GitHub repository.
  • [COMMAND_EXECUTION]: The skill relies on executing various subcommands of the rhwp native binary (e.g., info, dump, export-svg, convert) to process HWP files. This includes functionality to remove read-only locks from HWP documents.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted HWP/HWPX documents provided by the user to generate IR dumps, pagination reports, and SVG renders. Maliciously crafted documents could theoretically exploit the underlying parser or cause confusion when the agent processes the resulting textual dumps. The skill lacks explicit sanitization or boundary markers for these data ingestion points.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 01:41 AM
Security Audit — agent-trust-hub — rhwp-advanced