X Engagement Pro

Automates authentic engagement on X by monitoring AI image generation conversations, responding, amplifying content, and tracking metrics for brand growth.

Before installing, verify the following: 1) Resolve the metadata mismatch — the registry summary claims no required env vars but skill.json/README/agent code require X_API_KEY (and optional ALIGNED_NEWS_API_KEY). Only proceed if the registry/package owner confirms this. 2) Inspect the referenced GitHub repo (skill.json/README point to github.com/notphilsledge/x-engagement-pro); confirm the repo and author identity and review recent commits. 3) Confirm the 'xapi' CLI the code execs is a trusted binary: verify its origin, that it's the expected X API client, and that your PATH can't be manipulated to point to a malicious replacement. 4) Limit the X API key's permissions to the minimum (prefer read/write scopes as required but avoid broad account scopes), and consider creating a dedicated account for automated engagement. 5) Run the skill in manual mode first (no auto-post) and in an isolated/test gateway to observe behavior and logs, confirming it does not leak credentials or call unexpected endpoints. 6) If you rely on the OpenClaw runtime's openclaw-tools.exec implementation, confirm whether it logs environment variables or passes them to other systems. 7) If you are not comfortable with these open questions (source verification, CLI trust, metadata inconsistency), treat the skill as untrusted until you can audit the repository and runtime environment.

Type: OpenClaw Skill Name: x-engagement-pro Version: 1.2.0 The skill contains critical shell injection vulnerabilities in `agents/engagement-agent.js` where user-defined keywords and external data from the X API (post IDs) are directly interpolated into shell commands via the `exec` function (e.g., `xapi search --query "${kw}"` and `xapi reply --id ${post.id}`). While these flaws could allow for Remote Code Execution (RCE) if an attacker influences the input, there is no evidence of intentional malice, data exfiltration, or backdoors. The code's logic remains consistent with its stated purpose of automating social media engagement.