skills/openclaw/skills/boof/Gen Agent Trust Hub

boof

Pass

Audited by Gen Agent Trust Hub on Mar 9, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION]: Vulnerable to indirect prompt injection via document processing.\n
  • Ingestion points: The scripts/boof.sh script processes PDF files for conversion and indexing.\n
  • Boundary markers: Absent. No delimiters are used for document excerpts to distinguish them from system instructions.\n
  • Capability inventory: Local execution of processing tools and file system access for storage.\n
  • Sanitization: Extracted document content is not sanitized for malicious instructions or prompt-altering sequences.\n- [EXTERNAL_DOWNLOADS]: Documentation directs users to install dependencies from external sources like PyPI and GitHub (marker-pdf and tobi/qmd). These are legitimate requirements for the skill's conversion and indexing capabilities.\n- [COMMAND_EXECUTION]: The boof.sh script executes local binaries for conversion and indexing. It uses safe variable quoting and path handling to mitigate command injection risks from file names.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 9, 2026, 08:36 PM