skills/openai/plugins/writing-skills/Gen Agent Trust Hub

writing-skills

Pass

Audited by Gen Agent Trust Hub on Aug 26, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [Command Execution]: The skill includes a utility script, 'render-graphs.js', which executes system commands to process data. The script uses 'child_process.execSync' to run the 'dot' command from the Graphviz suite, enabling the conversion of markdown diagram blocks into SVG files. It performs an environment check using 'which dot' to ensure the necessary binary is installed before attempting to render diagrams.
  • [External References]: The documentation contains references to external websites and specifications for informational purposes. It links to the agent skills specification at 'agentskills.io' and provides guidance based on official best practices. These references are intended to support the user in following standardized authoring patterns and do not involve automated execution or data exfiltration.
  • [Dynamic Syntax Documentation]: The skill provides documentation regarding the use of dynamic context syntax supported by certain agent platforms. It mentions the '!command' pattern as a feature for executing shell commands at load time for tasks such as retrieving logs or status updates. This information is provided as architectural guidance for developers and does not contain malicious injection patterns.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 26, 2026, 11:02 PM
Security Audit — agent-trust-hub — writing-skills