ClawHub

X Management

v0.1.1

Draft-first X/Twitter account management workflow for OpenClaw. Use when the user wants to read their X account context, draft tweets, replies, quote tweets,...

⭐ 0· 87·0 current·0 all-time
by@princebansal

Install

OpenClaw Prompt Flow

Install with OpenClaw

Best for remote or guided setup. Copy the exact prompt, then paste it into OpenClaw for princebansal/x-management.

Previewing Install & Setup.
Prompt PreviewInstall & Setup
Install the skill "X Management" (princebansal/x-management) from ClawHub.
Skill page: https://clawhub.ai/princebansal/x-management
Keep the work scoped to this skill only.
After install, inspect the skill metadata and help me finish setup.
Use only the metadata you can verify from ClawHub; do not invent missing requirements.
Ask before making any broader environment changes.

Command Line

CLI Commands

Use the direct CLI path if you want to install manually and keep every step visible.

OpenClaw CLI

Bare skill slug

openclaw skills install x-management

ClawHub CLI

Package manager switcher

npx clawhub@latest install x-management
Security Scan
Capability signals
Posts externally
These labels describe what authority the skill may exercise. They are separate from suspicious or malicious moderation verdicts.
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Benign
high confidence
✓
Purpose & Capability
Name/description match the instructions: the skill is a workflow layer for an X plugin and does not request unrelated binaries, env vars, or config paths.
✓
Instruction Scope
SKILL.md limits actions to reading account context, creating drafts, requiring explicit approval, and publishing via the plugin API surface; it does not instruct reading local files, environment variables, or sending data to external endpoints outside the plugin.
✓
Install Mechanism
No install spec and no code files are included (instruction-only), so nothing will be written to disk or downloaded by the skill itself.
✓
Credentials
The skill declares no required environment variables, credentials, or config paths; its stated behavior (workflow guidance for the X plugin) does not justify additional secrets.
✓
Persistence & Privilege
Skill is user-invocable and not forced always-on; model invocation is permitted (default) which is normal for skills — there are no indications it seeks elevated or persistent system privileges.
Assessment
This skill is an agent-side, instruction-only workflow that expects you to also install and trust the complementary OpenClaw X plugin (the plugin will handle auth and publishing). It does not ask for credentials or install code itself, so the immediate risk from the skill text is low. Before using, confirm the X plugin's permissions and review its publish/auth flow — the plugin (not this skill) will require your account tokens and can perform writes. If you want extra caution, keep the agent's autonomous actions limited or require explicit confirmation for any publish operation so nothing is posted without your approval.

Like a lobster shell, security has layers — review code before you run it.

latestvk97447wdxe19y5tjw69yhd2c6h858zre
87downloads
0stars
2versions
Updated 6d ago
v0.1.1
MIT-0
@princebansal
latest
Download

X Management

Use the X plugin as a draft-first system, not an autonomous posting engine.

Core operating rule

Never treat a generated tweet/reply/quote as ready to publish by default. Always:

  1. gather context first when relevant
  2. create a draft
  3. show the draft clearly
  4. require explicit user approval
  5. only publish if approval has been captured through the plugin flow

Current real plugin capabilities

Working now:

  • x_account_connect
  • x_account_auth_url
  • x_account_complete
  • x_account_me
  • x_timeline_me
  • x_timeline_mentions
  • x_post_get
  • x_post_context
  • x_post_create
  • x_post_reply
  • x_post_quote
  • x_post_thread
  • x_post_approve
  • x_post_publish for approved single posts and approved thread drafts
  • x_media_upload
  • x_util_resolve_url

Still partial:

  • engagement actions
  • deeper thread expansion beyond immediate referenced context
  • packaged-runtime validation should still be treated as an active release-quality concern when new plugin versions ship

Use the plugin surface this way

Reading context

Prefer read actions before drafting whenever the content depends on an external post or thread. Use:

  • x_util_resolve_url to normalize a target post URL
  • x_post_get for a single post with normalized output
  • x_post_context for a post plus immediate referenced context
  • x_timeline_mentions when deciding whether/how to reply
  • x_timeline_me when recent account voice/context matters

Drafting

Use:

  • x_post_create for a normal tweet
  • x_post_reply for a reply
  • x_post_quote for a quote tweet
  • x_post_thread for multi-post threads

Treat returned draftId as the durable handle for follow-up approval/publish steps. For threads, keep each post concise and make sure the full sequence is approved before publishing.

Approval and publish

Use x_post_approve only after the user has explicitly approved the exact draft. Then use x_post_publish only for the approved draft id. x_post_publish can now publish approved single posts, replies, quotes, and thread drafts. Do not mutate draft text between approval and publish.

Drafting heuristics

For replies

Before drafting a reply:

  • inspect the target post
  • inspect immediate referenced context when available
  • inspect your recent voice if tone consistency matters
  • avoid replying to a hallucinated interpretation of the post
  • keep factual claims anchored to what is actually visible

For quote tweets

A quote tweet should add something:

  • analysis
  • framing
  • disagreement
  • emphasis
  • synthesis

Do not produce empty applause unless the user wants that tone.

For tweets

When the user does not specify tone, prefer concise, high-signal writing over bloated threadbait. Offer 2-3 options when tone uncertainty is real.

Safety boundaries

  • do not publish automatically
  • do not assume approval from conversational enthusiasm
  • do not like/repost/bookmark autonomously unless the user explicitly asks
  • do not DM from this workflow
  • if publish or read capabilities fail, explain the actual failure clearly

Current architectural split

  • plugin = auth, reads, drafts, approval state, publish primitive
  • agent/skill = judgment, research, tone, decision-making, proposing options

That is the intended shape for Prince’s eventual dedicated X-managing agent.

Comments

Loading comments...