Socialconductor Skill
ReviewAudited by ClawScan on May 10, 2026.
Overview
This skill’s purpose matches its behavior, but it can control OAuth-connected social media bots that may post, approve, or moderate public comments.
Install only if you are comfortable giving this workflow control over the intended social media accounts. Test in simulation mode, review drafts before bulk approval, use pause/disable commands when needed, and revoke OAuth/API access if you stop using the service.
Findings (4)
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.
A mistaken or overly broad command could post content publicly, approve many AI replies, or block a user on a connected channel.
The skill exposes commands that can publish or approve public replies and moderate users on connected social accounts.
`approve all pending drafts`; `reply to facebook comment abc123 saying Great question! | Posts manual reply`; `block @username | Adds @username to block list`; `facebook simulation mode off | Bot posts for real`
Use explicit prompts for posting, blocking, and bulk approvals; review logs or drafts first, and keep simulation or pause mode enabled while testing.
Connecting an account gives the SocialConductor/OpenClaw workflow authority to act on that social media channel, including reading comments and managing replies.
The skill requires connecting external social accounts through OAuth/API-style access so it can operate those accounts.
“Each platform uses its own API key” and “Open it, log in with Facebook ... All Facebook and Instagram commands are now active.”
Connect only accounts you intend to automate, prefer business/creator accounts with least necessary privileges, and revoke access from the platform dashboard if no longer needed.
Poorly chosen examples or copied comment text could persistently affect the bot’s future tone and responses.
The skill can store or use selected examples to influence future AI reply behavior.
“AI Teaching” ... “View "star" and "bad" example lists, and add new examples to train the AI tone.”
Add only reviewed examples, periodically audit teaching lists, and avoid saving sensitive or untrusted comment text as training guidance.
Once enabled, the bot may keep replying publicly until paused, disabled, or returned to simulation mode.
The skill can enable ongoing automated replies that continue after the initial command.
`enable facebook bot | Turns on auto-reply`; `resume my facebook bot | Hold OFF — AI replies resume`; `facebook simulation mode off | Bot posts for real`
Start in simulation mode, monitor early replies, and keep clear pause/disable procedures available for each connected platform.