Skills
Navigate
skills/reason-machines/devtools-skills/officecli-office-automation/Gen Agent Trust Hub

officecli-office-automation

Fail

Audited by Gen Agent Trust Hub on Sep 16, 2026

Risk Level: HIGHREMOTE_CODE_EXECUTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The skill provides installation instructions for macOS/Linux that pipe a remote script directly into the shell: curl -fsSL https://raw.githubusercontent.com/iOfficeAI/OfficeCLI/main/install.sh | bash.
  • [REMOTE_CODE_EXECUTION]: The skill provides installation instructions for Windows that execute remote PowerShell code directly: irm https://raw.githubusercontent.com/iOfficeAI/OfficeCLI/main/install.ps1 | iex.
  • [EXTERNAL_DOWNLOADS]: The skill downloads and installs a binary (officecli) from an untrusted GitHub organization (iOfficeAI).
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to read, parse, and view content from Office documents (Word, Excel, PowerPoint) using commands like officecli view document.docx text. This creates a vulnerability where instructions embedded within processed documents could potentially manipulate the agent's logic or downstream actions.
  • Ingestion points: Files processed by officecli view, officecli get, and officecli watch in SKILL.md.
  • Boundary markers: None identified in the prompt examples.
  • Capability inventory: Subprocess execution for officecli commands and shell scripts.
  • Sanitization: None identified; the skill assumes document content is safe to read and display.
Recommendations
  • HIGH: Downloads and executes remote code from: https://raw.githubusercontent.com/iOfficeAI/OfficeCLI/main/install.sh - DO NOT USE without thorough review
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Sep 16, 2026, 04:27 AM
Security Audit — agent-trust-hub — officecli-office-automation