Security audit
Security checks for vulnerabilities and agentic risk
This skill performs the advertised X/Twitter thread fetching and PDF saving, with disclosed network and browser use and only minor usage cautions.
Install only if you are comfortable sending the target tweet ID to twitter-thread.com. Use explicit requests such as reading or downloading a specific thread, and provide an output path for PDFs if you do not want files saved to Downloads by default.
The skill advertises shell-based operation (curl, python3, Chrome headless) but does not declare any explicit tool scope or allowed tools. That creates an authorization gap where an agent may invoke shell capabilities more broadly than intended, increasing the chance of command execution in contexts the user did not clearly authorize.
The trigger condition is broad enough to activate on any dropped X/Twitter URL in chat, which can cause the agent to fetch remote content without sufficiently specific user intent. In practice, this expands the attack surface for prompt-triggering, unintended network access, and automatic processing of attacker-supplied links.
The PDF mode writes a file to disk, defaulting to $HOME/Downloads/thread-.pdf when no output path is provided, without an explicit user-facing confirmation of the destination or the side effect. In an agent context, this can create unexpected persistence of fetched content on the local system, which is a real but low-severity safety issue rather than a direct code execution flaw.
No suspicious patterns detected.
