X CDP Automation

WarnAudited by ClawScan on May 10, 2026.

Overview

This skill is transparent about automating X, but it can use a persistent logged-in browser session to post publicly and advertises lower-detection browser automation.

Install only if you are comfortable giving the skill control of a logged-in X browser profile. Use a dedicated profile/account, approve every post before execution, prefer dry-run previews, close the CDP browser when finished, and consider pinning dependencies instead of auto-installing them to /tmp.

Findings (5)

Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.

What this means

If the skill or the exposed browser session is misused, it can post or publish as your X account.

Why it was flagged

The skill uses the user's logged-in browser session as its authority to act on X. That is broader than a scoped API token and means actions are performed as the user's account.

Skill content
Uses real browser sessions with existing login, no API keys needed. ... Log in to X ... The session persists in the profile directory.
Recommendation

Use a dedicated Chromium profile and, ideally, a dedicated X account; only run after reviewing the exact content; close the browser/CDP session when finished.

What this means

Using this for automation may violate platform rules, risk account enforcement, or enable spam-like behavior if not tightly controlled.

Why it was flagged

The skill frames lower detectability and human-like timing as benefits, which can encourage use that evades platform automation controls.

Skill content
**Detection**: CDP automation looks like real browser usage. Much harder to detect than API abuse. ... **Account safety**: Human-like delays are built into all scripts.
Recommendation

Do not use it for bulk, spam, or evasion; keep the approval step mandatory and follow X's platform rules.

What this means

A local process that can reach the debugging port may be able to control that browser session while it remains open.

Why it was flagged

Setup starts a detached Chromium instance with CDP enabled and a persistent profile, so the controllable browser can remain running after setup completes.

Skill content
`--remote-debugging-port=${args.port}`, `--user-data-dir=${args.profile}` ... `detached: true` ... `child.unref();`
Recommendation

Keep the CDP port local, close Chromium when done, and avoid reusing this profile for unrelated sensitive browsing.

What this means

The dependency code used to control the browser may change over time or be affected by npm supply-chain issues.

Why it was flagged

The setup can automatically install the latest puppeteer-core from npm into /tmp, without a pinned version or lockfile.

Skill content
execSync('cd /tmp && npm init -y 2>/dev/null; npm install puppeteer-core 2>&1', ...)
Recommendation

Prefer installing a pinned puppeteer-core version in a project directory you control, and review dependency provenance before use.

What this means

Accidental or premature execution could publish content publicly from your account.

Why it was flagged

The instructions include a user-approval gate, but the executed tools perform public account actions once run.

Skill content
Agent drafts the text, shows it to user for approval ... On confirmation, run the script via `exec`
Recommendation

Require explicit confirmation for every post/reply/article and use --dry-run to preview when possible.