walkthrough
Pass
Audited by Gen Agent Trust Hub on Nov 18, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [SAFE]: The skill implements several security best practices, including a redaction engine in
runtime/scripts/core.mjsthat identifies and masks sensitive credentials (API keys, tokens, private keys) before they are included in any generated output. - [SAFE]: The file collection logic explicitly excludes sensitive directories and file types, such as
.git,node_modules,.envfiles, and various certificate/key formats (.pem,.key, etc.), preventing accidental data exposure. - [SAFE]: Path validation in
runtime/scripts/core.mjsuses acheckedPathfunction to prevent directory traversal attacks by ensuring all accessed files remain within the target repository boundary. - [EXTERNAL_DOWNLOADS]: During its initial setup, the skill uses
npmto install required rendering dependencies (such as Shiki for syntax highlighting and Mermaid for diagrams) and uses Playwright to install a headless Chromium browser. These downloads are performed from official registries and are required for the skill's core functionality. - [COMMAND_EXECUTION]: The skill executes standard CLI tools including
git,gh(GitHub CLI), andffmpeg. These are used to capture source code context, interact with GitHub pull requests for delivery, and encode video artifacts, respectively. Subprocess calls are handled with timeouts and restricted execution environments.
Audit Metadata