安全扫描（ClawHub）

• VirusTotal ：良性
• OpenClaw ：良性

OpenClaw 评估

The skill is an instruction-only ‘human handoff’ helper for ad platforms and its requirements and instructions are internally consistent, but the package has no provenance (no homepage or source) so you should vet what account data you pass into it before use.

安装（复制给龙虾 AI）

将下方整段复制到龙虾中文库对话中，由龙虾按 SKILL.md 完成安装。

请把本段交给龙虾中文库（龙虾 AI）执行：为本机安装 OpenClaw 技能「Ads Human Handoff」。简介：Escalate automation conversations to human ad experts for Meta (Facebook/Instag…。
请 fetch 以下地址读取 SKILL.md 并按文档完成安装：https://raw.githubusercontent.com/openclaw/skills/refs/heads/main/skills/danyangliu-sandwichlab/human-handoff-coordinator/SKILL.md
（来源：yingzhi8.cn 技能库）

SKILL.md

打开原始 SKILL.md（GitHub raw）

---
name: human-handoff-coordinator
description: Escalate automation conversations to human ad experts for Meta (Facebook/Instagram), Google Ads, TikTok Ads, and YouTube Ads operations.
---

# Ads Human Handoff

## Purpose
Core mission:
- handoff packet creation, escalation routing

This skill is specialized for advertising workflows and should output actionable plans rather than generic advice.

## When To Trigger
Use this skill when the user asks for:
- ad execution guidance tied to business outcomes
- growth decisions involving revenue, roas, cpa, or budget efficiency
- platform-level actions for: Meta (Facebook/Instagram), Google Ads, TikTok Ads, YouTube Ads
- this specific capability: handoff packet creation, escalation routing

High-signal keywords:
- ads, advertising, campaign, growth, revenue, profit
- roas, cpa, roi, budget, bidding, traffic, conversion, funnel
- meta, googleads, tiktokads, youtubeads, amazonads, shopifyads, dsp

## Input Contract
Required:
- question: user issue or decision request
- context: account, campaign, and objective context
- urgency_level

Optional:
- error_message
- screenshots_or_logs
- preferred_response_style

## Output Contract
1. Direct Answer
2. Root Cause Hypothesis
3. Immediate Actions
4. Escalation Criteria
5. Follow-up Questions

## Workflow
1. Classify question type (how-to, diagnosis, policy, strategy).
2. Provide shortest valid answer first.
3. Add context-aware action checklist.
4. Flag escalation if risk or uncertainty is high.
5. Return follow-up fields only if required.

## Decision Rules
- If answer confidence is low, state uncertainty and propose verification steps.
- If issue impacts spend safety, prioritize pause or cap recommendations.
- If user asks unsupported action, hand off with exact context package.

## Platform Notes
Primary scope:
- Meta (Facebook/Instagram), Google Ads, TikTok Ads, YouTube Ads

Platform behavior guidance:
- Keep recommendations channel-aware; do not collapse all channels into one generic plan.
- For Meta and TikTok Ads, prioritize creative testing cadence.
- For Google Ads and Amazon Ads, prioritize demand-capture and query/listing intent.
- For DSP/programmatic, prioritize audience control and frequency governance.

## Constraints And Guardrails
- Never fabricate metrics or policy outcomes.
- Separate observed facts from assumptions.
- Use measurable language for each proposed action.
- Include at least one rollback or stop-loss condition when spend risk exists.

## Failure Handling And Escalation
- If critical inputs are missing, ask for only the minimum required fields.
- If platform constraints conflict, show trade-offs and a safe default.
- If confidence is low, mark it explicitly and provide a validation checklist.
- If high-risk issues appear (policy, billing, tracking breakage), escalate with a structured handoff payload.

## Code Examples
### Quick Triage JSON

    {
      "issue_type": "delivery_drop",
      "severity": "medium",
      "first_actions": ["check spend cap", "check policy status"]
    }

### Handoff Payload

    ticket_type: platform_support
    required_fields: [account_id, campaign_id, timeline, last_change]

## Examples
### Example 1: Delivery suddenly dropped
Input:
- Campaign impressions down 60%
- No recent manual changes

Output focus:
- probable causes
- first 3 checks
- escalation trigger

### Example 2: Policy rejection question
Input:
- Ad rejected with vague reason
- User wants fastest fix

Output focus:
- policy interpretation
- rewrite direction
- approval retry order

### Example 3: Need human support now
Input:
- Billing or account lock issue
- Launch deadline is today

Output focus:
- handoff packet
- urgency level
- required owner and ETA

## Quality Checklist
- [ ] Required sections are complete and non-empty
- [ ] Trigger keywords include at least 3 registry terms
- [ ] Input and output contracts are operationally testable
- [ ] Workflow and decision rules are capability-specific
- [ ] Platform references are explicit and concrete
- [ ] At least 3 practical examples are included