human-handoff-coordinator

What This Skill Does

The human-handoff-coordinator is a specialized OpenClaw skill designed to bridge the gap between automated AI agents and human ad experts. Its primary purpose is to ensure that complex, high-stakes advertising decisions regarding Meta (Facebook/Instagram), Google Ads, TikTok Ads, and YouTube Ads are managed with professional oversight. Rather than providing generic advice, the skill generates a structured, actionable handoff packet that includes root cause hypotheses, immediate tactical actions, and clear escalation criteria for human intervention.

Installation

To install this skill, run the following command in your terminal within the OpenClaw environment: clawhub install openclaw/skills/skills/danyangliu-sandwichlab/human-handoff-coordinator Ensure that you have the necessary environment permissions to add skills from the openclaw/skills repository.

Use Cases

This skill is indispensable for performance marketers and agency teams who need to scale operations safely. It is triggered during scenarios such as:

• Sudden delivery drops in Meta campaigns requiring urgent creative audit.
• Complex bidding strategy adjustments for Google Ads that impact ROAS and revenue.
• Policy-related issues or account-level flags that threaten spend efficiency.
• Strategic pivots requiring a combination of platform-level actions and human approval.

Example Prompts

1. "My Google Ads campaign ROAS has dropped by 40% in the last 48 hours. Can you analyze the search terms and prepare a handoff packet for my account manager?"
2. "We need to increase our TikTok Ads budget by $5k/day, but I'm worried about audience saturation. Create a strategy assessment and escalation plan for the creative team."
3. "Facebook ads were rejected due to policy violations. Review the error code and generate the necessary details for human support to contest this."

Tips & Limitations

The human-handoff-coordinator excels at structuring information but is not a replacement for human judgment in gray-area policy enforcement or creative branding nuances. It is designed to prioritize safety; always pay close attention to the 'rollback' and 'stop-loss' conditions generated by the skill when dealing with high-spend scenarios. When the skill identifies a low confidence level, do not attempt to automate the final execution; treat the output as a draft for human review. Ensure your context inputs (account IDs and campaign objectives) are accurate, as the quality of the escalation payload depends entirely on the precision of the data provided.