← Back to Skills Marketplace
527
Downloads
1
Stars
0
Active Installs
11
Versions
Install in OpenClaw
/install x-twitter-browser
Description
Log in to X/Twitter via a real browser session and perform actions — posting, replying, reposting, liking, and bookmarking tweets via headless Playwright.
Usage Guidance
This skill appears to do what it claims: it automates X by saving and reusing a browser session. Before installing, consider the following: (1) Session cookies saved at ~/.openclaw/auth/x-twitter/cookies.json are sensitive — treat them like credentials. Use a dedicated/testing X account rather than a high-value account. (2) Ensure filesystem permissions on ~/.openclaw/auth restrict access (other local users/skills could read cookies). (3) The scripts include anti-detection tweaks (navigator.webdriver/window.chrome) — expected for automation but may violate X/Twitter terms of service; be aware of account policy risk. (4) The setup installs Playwright and Chromium from standard tooling; you can run setup locally and then copy the cookies to a headless server (the SKILL.md documents this) if you don't want to perform login on the server. (5) The login flow relies on the agent not pressing Enter until you confirm — if you distrust autonomous invocation, avoid granting this skill automatic agent invocation or run the login manually. If you need higher assurance, inspect or run the scripts locally before adding the skill to OpenClaw.
Capability Analysis
Type: OpenClaw Skill
Name: x-twitter-browser
Version: 1.0.10
The x-twitter-browser skill is a legitimate tool for automating X/Twitter actions via Playwright browser sessions. It manages authentication by storing session cookies locally in a centralized directory (~/.openclaw/auth/x-twitter/cookies.json) and provides scripts for posting, replying, liking, and bookmarking. The code follows standard browser automation practices, including anti-detection measures in session_lib.py, and the SKILL.md instructions correctly guide the agent through a manual login process without any evidence of malicious intent or data exfiltration.
Capability Assessment
Purpose & Capability
Name/description, SKILL.md, and the included Python scripts all implement headless/browser Playwright automation for X (post, reply, repost, like, bookmark). Required binary is only python3 and the setup installs Playwright + Chromium — all expected for the stated purpose. There are no unrelated environment variables or third-party credentials requested.
Instruction Scope
Runtime instructions are explicit and limited to logging in via a visible browser, saving Playwright storage_state to ~/.openclaw/auth/x-twitter/cookies.json, verifying sessions, and running the action scripts. The SKILL.md instructs the agent to wait for user confirmation before sending Enter to save cookies — this flow is appropriate but relies on the agent following the prompt. The skill manipulates browser characteristics (navigator.webdriver, window.chrome) to mask automation; that's consistent with avoiding detection but worth noting as evasion behavior.
Install Mechanism
No external or opaque downloads: setup.sh installs Playwright via pip and runs 'python3 -m playwright install chromium' (official Playwright tooling). Chromium is downloaded via Playwright's standard mechanism. No extract-from-arbitrary-URL steps or unknown hosts are used.
Credentials
The skill requests no environment variables or API keys. It does, however, create and rely on a saved cookie file (~/.openclaw/auth/x-twitter/cookies.json). Those cookies are effectively account credentials and are sensitive — their presence is proportionate to the purpose but important to treat as secrets (do not share, commit, or expose them).
Persistence & Privilege
always is false (no forced global inclusion). The skill persists session cookies into OpenClaw's centralized auth directory, which survives updates. That is normal for this use-case but means other skills or processes with access to ~/.openclaw/auth could read those cookies if they have permission — consider filesystem permissions and trust of other installed skills.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install x-twitter-browser - After installation, invoke the skill by name or use
/x-twitter-browser - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.10
- Improved session login flow: Now uses a two-phase "agent-driven" process where Enter is sent to the script only after the user confirms login, rather than requiring direct terminal interaction by the user.
- Updated documentation to clarify operational requirements for session setup and management.
- Added detailed instructions for handling login prompts and warnings during session creation.
- Clarified workflow: agent now fully orchestrates session setup, enhancing user guidance and automation.
v1.0.9
**Major change: Migrate from manual cookie headers to persistent browser login sessions with improved session management.**
- Added `scripts/setup_session.py` for session login via a visible browser window and centralized persistent cookie storage in `~/.openclaw/auth/x-twitter/cookies.json`.
- Removed `scripts/save_cookie_header.py` and all manual cookie header workflows.
- All browser automation now uses the saved login session; no need to supply headers for each use.
- Session setup, verification, and troubleshooting instructions are updated for the new workflow.
- Existing tweet posting, replying, (re)posting, liking, and bookmarking features remain, now using the persistent auth model.
- Documentation updated to reflect new session management and centralized config.
v1.0.8
No file changes were detected in this version.
- No code or documentation updates.
- Version and features remain unchanged.
v1.0.7
No changes detected in this version.
- Version number updated to 1.0.7, but all files remain unchanged from the previous release.
v1.0.6
Added liking and bookmarking tweet actions.
- Added support to like and unlike tweets via scripts/like_post.py
- Added support to bookmark and remove bookmark from tweets via scripts/bookmark_post.py
- Updated documentation to include like and bookmark workflows and commands
v1.0.5
- Session cookie storage moved from skill-local directory (runtime/cookie-header.txt) to global user config (~/.x-twitter-browser/config.json) for persistence across skill updates.
- Documentation updated to reflect new config location and usage instructions.
- Local runtime directory is no longer used for cookie or config storage; cookie header and future settings are saved in a global JSON config file.
- All operational instructions and examples now reference ~/.x-twitter-browser/config.json.
v1.0.4
Added reply and repost capabilities.
- New: Added reply_post.py to allow replying to tweets by URL or ID.
- New: Added repost_post.py to allow retweeting and quoting tweets by URL or ID.
- Updated README/instructions in SKILL.md to describe replying and reposting features.
- Skill description updated to mention support for posting, replying, and reposting tweets.
v1.0.3
No user-facing changes in version 1.0.3; documentation and file structure remain unchanged.
v1.0.2
- Added scripts/setup.sh to provide a convenient one-step setup for installing dependencies.
- Updated documentation to recommend first-time setup using scripts/setup.sh for easier installation.
v1.0.1
- README (SKILL.md) fully rewritten in concise English for clarity and brevity.
- Documentation now focuses only on current features: session setup, login state verification, and tweet posting using a cookie header.
- References to Playwright storageState and future features (reply, like, etc.) removed for tighter scope.
- All instructions streamlined for simplicity and quick setup.
- Sections on workflow, dependencies, troubleshooting, and user warnings made more direct and beginner-friendly.
v1.0.0
x-twitter-browser 1.0.0
- 首发版本，实现基于真实浏览器 cookie 或 Playwright storageState 的 X/Twitter 登录验证与浏览器自动化。
- 支持通过 cookie header 或 storageState 生成和复用浏览器会话。
- 已实现自动化发帖（纯文本）。
- 提供会话有效性验证（--verify-only）功能。
- 详细文档指导 cookie/header 获取方法、依赖安装与运行流程。
- 设计便于后续扩展到回复、点赞、转推、关注等动作。
Metadata
Frequently Asked Questions
What is x-twitter-browser?
Log in to X/Twitter via a real browser session and perform actions — posting, replying, reposting, liking, and bookmarking tweets via headless Playwright. It is an AI Agent Skill for Claude Code / OpenClaw, with 527 downloads so far.
How do I install x-twitter-browser?
Run "/install x-twitter-browser" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is x-twitter-browser free?
Yes, x-twitter-browser is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does x-twitter-browser support?
x-twitter-browser is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created x-twitter-browser?
It is built and maintained by clawgrid (@zhanghanjun2); the current version is v1.0.10.
More Skills