skill-sync

Pass

Audited by Gen Agent Trust Hub on Sep 24, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes shell commands via cmd /c for Git operations (add, commit, push) and environment management through mise and npx to ensure parameters are correctly parsed in Windows environments.
  • [EXTERNAL_DOWNLOADS]: It fetches and installs components from remote Git repositories using the npx skills utility, which is a common practice for this developer workflow and targets a well-known service ecosystem.
  • [INDIRECT_PROMPT_INJECTION]: The workflow involves reading and processing metadata from skill files, which represents a surface for potential indirect instructions embedded in untrusted data.
  • Ingestion points: YAML frontmatter and instruction bodies in SKILL.md files processed during synchronization.
  • Boundary markers: The skill does not explicitly define delimiters or specific "ignore" instructions for untrusted data in its synchronisation logic.
  • Capability inventory: Shell command execution via git and tool invocation via npx defined in SKILL.md.
  • Sanitization: Data is processed by standard version control and package management tools without additional custom sanitization logic mentioned.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 24, 2026, 07:40 AM