officecli-word-form

Fail

Audited by Gen Agent Trust Hub on Aug 28, 2026

Risk Level: HIGHREMOTE_CODE_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The skill provides instructions to install the 'officecli' tool by downloading scripts from 'https://d.officecli.ai/install.sh' (Linux/macOS) and 'https://d.officecli.ai/install.ps1' (Windows) and piping them directly to shell interpreters ('bash' and 'iex'). This allows the execution of remote code with the user's privileges without inspection.
  • [DYNAMIC_EXECUTION]: The skill uses 'python3 -c' to execute a multi-line script for processing OpenXML content. This involves dynamic script generation where shell variables derived from document metadata are interpolated into the Python environment.
  • [INDIRECT_PROMPT_INJECTION]:
  • Ingestion points: Untrusted data enters the agent context through 'officecli raw', 'query', and 'view' commands that read external Word documents (SKILL.md).
  • Boundary markers: There are no delimiters or 'ignore instructions' warnings provided when processing document content.
  • Capability inventory: The skill possesses capabilities for file modification, shell command execution, and remote script installation (SKILL.md).
  • Sanitization: There is no evidence of validation or escaping for document content before it is processed by shell utilities or Python scripts.
Recommendations
  • HIGH: Downloads and executes remote code from: https://d.officecli.ai/install.sh - DO NOT USE without thorough review
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Aug 28, 2026, 08:33 AM
Security Audit — agent-trust-hub — officecli-word-form