shortlink-qr
Warn
Audited by Gen Agent Trust Hub on Jun 18, 2026
Risk Level: MEDIUMCOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes multiple shell-level commands, including Git operations (
git commit,git fetch,git reset) and interactions with theportlessCLI tool. It specifically manages code checkouts and commits changes to a repository. - [DATA_EXFILTRATION]: The skill is configured to copy generated image assets directly to the macOS clipboard by default (
clipboard: true). This behavior can lead to unintentional data exposure or the placement of malicious content into the user's clipboard buffer. - [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it ingests user-supplied data (such as
slug,url, andtitle) and interpolates them directly into local configuration files (_redirects) and markdown files (shortlinks.svx) without any specified sanitization, escaping, or validation protocols. - [DATA_EXPOSURE]: The skill interacts with absolute file paths within the user's home directory (
/Users/joel/Code/joelhooks/...). Hardcoding sensitive local paths exposes the internal directory structure and increases the risk of unauthorized file modification if the agent's logic is subverted.
Audit Metadata