skills/daaab/openclaw-skills/nano-pdf/Gen Agent Trust Hub

nano-pdf

Pass

Audited by Gen Agent Trust Hub on Mar 9, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill installs the nano-pdf package from the public Python Package Index (PyPI) during the installation phase.
  • [COMMAND_EXECUTION]: The skill executes the nano-pdf command-line tool to perform file system operations, specifically editing PDF documents.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it transforms natural language instructions into document modifications. If an attacker provides these instructions via untrusted data (e.g., a summarized email or web page), they could potentially manipulate the resulting PDF output.
  • Ingestion points: The instruction string passed to the nano-pdf edit command.
  • Boundary markers: None identified in the skill definition.
  • Capability inventory: Modifies local PDF files via the nano-pdf CLI.
  • Sanitization: Not explicitly implemented in the skill wrapper.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 9, 2026, 10:18 AM