vault-cleanup-auditor

Pass

Audited by Gen Agent Trust Hub on Nov 16, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes bash scripts involving 'find', 'grep', 'wc', and 'stat' to perform filesystem audits and generate reports within the user's vault.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes file paths and file content from a local vault, which serves as a surface for indirect prompt injection.
  • Ingestion points: Filenames, directory structures, and Markdown content (specifically looking for status tags and headers) within the user-provided vault path (SKILL.md).
  • Boundary markers: None. The skill does not implement specific delimiters or instructions to the agent to disregard instructions potentially embedded in the files being analyzed.
  • Capability inventory: Bash tool access allows for file discovery, content searching, directory creation, and writing reports to the filesystem (SKILL.md).
  • Sanitization: No input validation or sanitization is performed on the user-provided vault path or the file metadata and content retrieved during the auditing process.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 16, 2026, 08:10 AM