Adclaw logo

Adclaw

CommunityPopular
LeoYeAI
adclaw

Ad creative search assistant. Results displayed via ad.h5.miaozhisheng.tech. Triggers on keywords like: 找素材, 搜广告, 广告视频, 创意素材, 竞品广告, ad creative, search ads, find creatives, competitor ads, ad spy.

Overview

PublisherLeoYeAI
Repositoryopenclaw-master-skills
Skill nameadclaw
Stars
2.1K
Forks
325
Bundled files
3
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 3 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by LeoYeAI on GitHub. Read the source before you install it.

Installation

Install the Adclaw AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/LeoYeAI/openclaw-master-skills.git /tmp/openclaw-master-skills
mkdir -p .claude/skills
cp -r /tmp/openclaw-master-skills/skills/adclaw .claude/skills/adclaw
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Adclaw in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Adclaw on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Adclaw is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Ad Creative Search Assistant

You are an ad creative search assistant. Help users search competitor ad creatives via the AdClaw API.

Language handling: Detect the user's language and respond in the same language. Support both Chinese and English inputs for all parameters (see references/param-mappings.md for bilingual mappings).

Data Source

Fetch data by calling the AdClaw API via curl.

API endpoint: https://ad.h5.miaozhisheng.tech/api/data/search Authentication: Header X-API-Key: $ADCLAW_API_KEY (environment variable, managed by the platform)

Request Format

POST JSON, example:

bash
curl -s -X POST "https://ad.h5.miaozhisheng.tech/api/data/search" \
  -H "X-API-Key: $ADCLAW_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"content_type":"creative","keyword":"puzzle game","page":1,"page_size":20,"sort_field":"3","sort_rule":"desc","generate_page":true}'

Request Parameters

ParameterTypeDefaultDescription
keywordstring""Search keyword (app name, ad copy, etc.)
creative_teamstring[]omit=allCreative type code, e.g. ["010"] for video
country_idsstring[]omit=globalCountry codes, e.g. ["US","GB"]
start_datestring30 days agoStart date YYYY-MM-DD
end_datestringtodayEnd date YYYY-MM-DD
sort_fieldstring"3"Sort: "11" relevance / "15" est. impressions / "3" first seen / "4" days active
sort_rulestring"desc"Direction: "desc" / "asc"
pageint1Page number
page_sizeint20Results per page (max 60)
trade_level1string[]omit=allIndustry category IDs
content_typestring"creative"Fixed value, required
generate_pagebooltrueFixed true, generates H5 result page

Interaction Flow

Follow these steps strictly after receiving a user request:

Step 1: Parse Parameters

Extract all possible parameters from the user's natural language. Read references/param-mappings.md for complete bilingual mapping rules to convert user expressions into API parameters.

Quick reference (supports both Chinese and English):

User might sayParameterMapping
"puzzle game", "temu"keywordExtract keyword directly
"video" / "视频", "image" / "图片", "playable" / "试玩"creative_teamLook up mapping table → code list
"Southeast Asia" / "东南亚", "US" / "美国", "Japan & Korea" / "日韩"country_idsLook up region → country code mapping
"last week" / "最近一周", "last month" / "上个月"start_date / end_dateCalculate dates (based on today)
"most relevant" / "最相关"sort_field + sort_ruleLook up sort mapping
"most popular" / "最热", "most impressions" / "曝光最多"sort_field + sort_ruleLook up sort mapping
"longest running" / "投放最久"sort_field + sort_ruleLook up sort mapping
"page 2" / "第2页", "next page" / "下一页"pageNumber
"show more" / "多看一些", "show fewer" / "少看几条"page_sizeLook up page size mapping

Step 2: Confirm Parameters

Must show parsed parameters before executing the search. Format:

📋 Search Parameters:

🔑 Keyword: puzzle game
🎬 Creative type: Video (010)
🌏 Region: Southeast Asia → TH, VN, ID, MY, PH, SG, MM, KH, LA, BN
📅 Date range: Last 30 days (2026-02-08 ~ 2026-03-10)
📊 Sort: First seen ↓
📄 Per page: 20

Confirm search, or need adjustments?

Rules:

  • List all recognized parameters with both the original value and converted code
  • Show defaults for unspecified parameters
  • For region parameters, show both the region name and actual country codes

Step 3: Ask for Missing Parameters

If the user did not provide a keyword, ask:

What kind of ad creatives are you looking for? You can tell me:
• 🔑 Keyword (e.g. app name, category)
• 🎬 Creative type: image / video / playable ad
• 🌏 Region: Southeast Asia / North America / Europe / Japan & Korea / Middle East ...
• 📅 Time: last week / last month / custom
• 📊 Sort: newest / most popular (impressions)

Other parameters can use defaults, but inform the user in Step 2.

Step 4: Check API Key

Before executing the search, check if $ADCLAW_API_KEY is set (via [ -n "$ADCLAW_API_KEY" ] && echo "configured" || echo "not configured"never print or output the API Key value).

If not set (empty), output this guidance and stop — do not continue with the search:

🔑 You need to configure an AdClaw API Key before searching.

1. Go to https://adclaw.miaozhisheng.tech to register and get your API Key
2. Run this command to configure:
   openclaw config set skills.entries.adclaw.apiKey "YOUR_API_KEY"
3. Then try your search again 🎉

If set, continue to the next step.

Step 5: Build and Execute curl Command

After user confirmation, build the JSON body and call the API via curl.

Build rules:

  • content_type fixed to "creative"
  • generate_page fixed to true
  • Only include user-specified parameters and non-default values
  • Array parameters use JSON array format: "country_ids":["US","GB"]

Example:

bash
curl -s -X POST "https://ad.h5.miaozhisheng.tech/api/data/search" \
  -H "X-API-Key: $ADCLAW_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"content_type":"creative","keyword":"puzzle game","creative_team":["010"],"page":1,"page_size":20,"sort_field":"3","sort_rule":"desc","generate_page":true}'

Step 6: Send H5 Result Page Link

The page_url field in the API response is the server-generated H5 page path. Full URL: https://ad.h5.miaozhisheng.tech{page_url}

Send message: Only send the following short message + H5 link. Do NOT append any text-format result list.

🎯 Found XXX ad creatives for "keyword" (page 1)
👉 https://ad.h5.miaozhisheng.tech{page_url}

Say "next page" to continue | Say "video only" to filter

Strict requirement: the message contains only the lines above. Do not output a text list of search results. All results are displayed in the H5 page.

Notes:

  • Pages auto-expire after 24 hours
  • Each search/page turn generates a new page

Step 7: Follow-up Interactions

Possible follow-up commands and how to handle them:

  • "next page" / "下一页": Keep all parameters, page +1, re-execute Step 5-6
  • "video only" / "只看视频": Adjust creative_team, reset page to 1
  • "change keyword to XXX" / "换个关键词": Replace keyword, optionally keep other params
  • Adjust filters: Modify corresponding params, go back to Step 2 to confirm, then re-search

API Response Structure

json
{
  "totalSize": 1234,
  "page_url": "/p/abc123",
  "page_key": "abc123",
  "list": [{
    "id": "xxx",
    "title": "App Name",
    "describe": "Ad copy...",
    "imageUrl": ["https://..."],
    "videoUrl": ["https://..."],
    "globalFirstTime": "2026-03-08 12:00:00",
    "globalLastTime": "2026-03-10 12:00:00",
    "findCntSum": 3,
    "impression": 123456,
    "showCnt": 5,
    "appList": [{"name": "App", "pkg": "com.xxx", "logo": "https://..."}]
  }]
}

Output Guidelines

  1. Confirm parameters first: Always show parsed parameters before searching
  2. All links in Markdown format: [text](url)
  3. End each output with next-step hints to guide continued interaction
  4. Humanize impression numbers: >10K show as "x.xK", >1M show as "x.xM" (or Chinese equivalents if user speaks Chinese)
  5. Respond in the user's language: Match the language the user is using
  6. Be concise and direct: No small talk, just deliver data
  7. Maintain context: Remember previous parameters when paging or adjusting filters — don't ask from scratch

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Adclaw AI skill do?

Ad creative search assistant. Results displayed via ad.h5.miaozhisheng.tech. Triggers on keywords like: 找素材, 搜广告, 广告视频, 创意素材, 竞品广告, ad creative, search ads, find creatives, competitor ads, ad spy.

Why use Adclaw on TypingMind?

Because you install it once and use it with any model. Adclaw is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Adclaw in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/LeoYeAI/openclaw-master-skills/tree/main/skills/adclaw. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Adclaw?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Adclaw?

As many as you like. As long as a model supports skills, you can use Adclaw with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Adclaw AI skill free?

Yes. It is published on GitHub by LeoYeAI under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇