pipiads（广告洞察）

PipiAds - TikTok & Facebook Adspy Intelligence

AI-powered TikTok & Facebook Adspy and Ad Library research for ads, products, stores, landing pages, advertisers, and competitors.

When to Use

Use this skill when the user wants to:

• Find trending or long-running TikTok or Facebook ads
• Research competitors, advertisers, stores, landing pages, or campaigns
• Discover winning products or fast-rising TikTok Shop items
• Analyze store performance, regional distribution, ad copy, delivery trends, creative hooks, or recurring ad angles
• Research ad trends across ecommerce, DTC, dropshipping, TikTok Shop, apps, games, drama, and AI tools
• Monitor advertisers or campaigns over time
• Search for ads, products, stores, or TikTok Shop items using an image

When NOT to Use

Do not use this skill when:

• The user only wants general web search results or news
• The user wants creative strategy without underlying ad intelligence data
• The user asks for analytics outside TikTok, Facebook, Meta Ad Library, or PipiAds-supported commerce data
• A broad query would waste credits when a narrower query or clarification is needed first

Setup

1. Visit the official website at https://pipiads.com/ to view real-time dashboard data and product UI.
2. Visit https://www.pipispy.com/ to get your API key and manage billing/recharge.
3. Set the environment variable: PIPIADS_API_KEY

Installation Note

This skill installs and runs the published npm package pipiads-mcp-server@1.0.3 as a local MCP server.

Default install:

npm install -g pipiads-mcp-server@1.0.3

If you prefer stricter isolation, use this skill in a dedicated development environment, container, or non-critical workstation.

Trust & Execution Model

This skill installs pipiads-mcp-server@1.0.3 from the npm registry and runs it locally as an MCP server.

This means:

• npm install scripts may run during installation
• the installed package executes on the local machine
• all API access is mediated through that local MCP server

If you require stricter controls, review the npm package contents and run it in an isolated environment before using it on a primary workstation.

Verification

This skill depends on:

• official website: https://pipiads.com/
• API portal: https://www.pipispy.com/
• npm package: pipiads-mcp-server@1.0.3

Review the package source, publisher, and release history if you need stronger supply-chain assurance before installation.

Data Handling

This skill sends search queries and optional image inputs to the external PipiAds service through the PipiAds API.

Do not send:

• sensitive personal data
• private customer information
• confidential campaign assets
• unnecessary image inputs

Use image search only when the provided image is safe to transmit to a third-party service.

Cost Awareness

Each API call consumes credits from your PipiAds account:

• List/Search: 1 credit per result returned
• Detail: 1 credit per request (free if queried within 3 days)

Prefer narrow queries before broad searches. If the user request is underspecified, clarify platform, region, time range, category, or target object before searching.

Core Workflow

1. Identify the target object: ad, product, TikTok Shop product, store, advertiser, app, or monitor task
2. Clarify the platform and scope when missing: TikTok, Facebook, Meta Ad Library, region, time range, category, price band, or performance threshold
3. Start with the relevant search/list tool to narrow the candidate set
4. Use detail or analysis tools to inspect the most relevant candidates
5. If the user wants ongoing tracking, use the monitor tools
6. If the user provides an image, use the AI image search workflow

Quick Reference

Best Practices

• Default to asking for platform if the user does not specify TikTok vs Facebook
• Ask for region and time range before broad searches whenever those constraints materially affect results
• Prefer search/list first, then detail, to avoid unnecessary credit usage
• For competitor analysis, identify the advertiser or store first, then drill into store analysis or monitor tools
• For monitoring operations, confirm the exact target if the user’s request is ambiguous
• For image search, always submit first, then poll status, then fetch summary/results
• When the user asks for “best” or “winning” ads/products, translate that into explicit metrics such as views, likes, duration, GMV, sales, or ranking filters

Available Tools (73 total)

Ad Spy

• search_ads — Search ad videos across TikTok and Facebook by keyword, region, platform, engagement metrics, ad spend, delivery days
• get_ad_detail — Get detailed info for a specific ad by video ID

Ad Products

• search_products — Search e-commerce products advertised on TikTok/Facebook
• get_product_detail — Get product detail including images, price, landing page, Shopify data
• search_adlibrary_products — Search Meta Ad Library products
• get_adlibrary_product_detail — Get Meta Ad Library product detail

TikTok Shop

• search_tiktok_products — Search TikTok Shop products by sales, GMV, price, trends
• get_tiktok_product_detail — Get TikTok Shop product detail
• search_tiktok_shops — Search TikTok Shop stores
• get_tiktok_shop_detail — Get TikTok Shop store detail

Advertisers & Stores

• search_advertisers — Search advertiser leaderboard
• get_advertiser_detail — Get advertiser detail
• search_stores — Search store ranking list
• get_store_detail — Get store detail by ID

Store Analysis (13 tools)

• get_store_competition — Competition analysis
• get_store_data_analysis — All-platform data analysis
• get_store_play_cost — Play cost statistics
• get_store_region_analysis — Regional ad distribution
• get_store_ad_copy_analysis — TikTok ad copy analysis
• get_store_ad_schedule — Ad campaign schedule
• get_store_product_analysis — Product analysis
• get_store_rank_data — Traffic ranking data
• get_store_ad_trend — Meta Ad Library ad trend
• get_store_delivery_analysis — Meta Ad Library delivery analysis
• get_store_longest_run_ads — Longest-running ad content
• get_store_most_used_ads — Most-used ad content
• get_store_fb_pages — Facebook advertiser pages

Meta Ad Library

• search_lib_ads — Search Meta Ad Library ads
• get_lib_ad_detail — Get ad detail by ID

Rankings

• get_product_ranking — Ad product ranking
• get_new_product_ranking — New product ranking
• get_app_ranking — Top apps ranking
• get_new_app_ranking — New apps ranking
• get_app_dev_ranking — App developers ranking

Apps

• search_apps — Search apps on TikTok/Facebook
• get_app_detail — Get app detail
• search_app_developers — Search app developers
• get_app_developer_detail — Get developer detail

Natural Traffic

• search_natural_videos — Search TikTok natural traffic videos

Ad Monitor (20 tools)

• search_fb_advertisers — Search Facebook advertisers for monitoring
• create_monitor_task — Create monitoring task
• list_monitor_tasks — List monitoring tasks
• get_monitor_task_detail — Task details
• cancel_monitor_task — Cancel task
• get_monitor_board — Dashboard overview
• set_monitor_task_group — Assign task to group
• get_monitor_realtime_overview — Real-time overview
• get_monitor_daily_overview — Daily overview
• get_monitor_landing_pages_overview — Landing pages overview
• get_monitor_latest_products — Latest products
• get_monitor_product_list — Product list
• get_monitor_landing_page_list — Landing page list
• get_monitor_ad_count_stats — Ad count statistics
• get_monitor_deep_analysis — Deep campaign analysis
• get_monitor_most_used_copy — Most-used ad copy
• get_monitor_longest_run_copy — Longest-running ad copy
• get_monitor_ad_list — Ad list
• get_monitor_ad_detail — Ad detail
• get_monitor_product_stats — Product statistics

Monitor Groups & Notifications

• create_monitor_group — Create group
• list_monitor_groups — List groups
• update_monitor_group — Update group
• delete_monitor_group — Delete group
• get_monitor_notifications — Get notification settings
• save_monitor_notifications — Save notification settings

AI Image Search (8 tools)

• ai_search_submit_image — Submit image for visual search
• ai_search_image_status — Check processing status
• ai_search_image_summary — Get result summary
• ai_search_image_ads — Search similar ads
• ai_search_image_products — Search similar products
• ai_search_image_stores — Search similar stores
• ai_search_image_tiktok_products — Search similar TikTok products
• ai_search_image_tiktok_shops — Search similar TikTok shops

Recommended Workflows

1. Ad Research Workflow

Use when the user wants to find winning creatives, trending campaigns, or competitor ads.

1. Start with search_ads
2. Narrow by keyword, region, platform, delivery days, spend, or engagement filters
3. Inspect the strongest candidates with get_ad_detail
4. Summarize patterns such as hooks, offer angles, duration, or cross-region distribution

2. Product Discovery Workflow

Use when the user wants advertised products or product validation.

1. Start with search_products or search_adlibrary_products
2. Filter by platform, views, timing, niche, or storefront characteristics
3. Use get_product_detail or get_adlibrary_product_detail for deeper analysis
4. Highlight pricing, landing page patterns, Shopify signals, and creative overlap

3. TikTok Shop Workflow

Use when the user asks for rising TikTok Shop products or stores.

1. Use search_tiktok_products or search_tiktok_shops
2. Narrow by sales, GMV, price, category, or growth trend
3. Use get_tiktok_product_detail or get_tiktok_shop_detail
4. Summarize momentum, pricing, and store/product positioning

4. Store / Competitor Analysis Workflow

Use when the user wants to understand a store’s advertising behavior.

1. Identify the store with search_stores or use a known store ID
2. Fetch baseline info with get_store_detail
3. Choose the right analysis tool based on the question:

- competition → get_store_competition - product mix → get_store_product_analysis - region mix → get_store_region_analysis - copy patterns → get_store_ad_copy_analysis - campaign timing → get_store_ad_schedule - trend / longest-run / reuse → corresponding store analysis tools

5. Meta Ad Library Workflow

Use when the user explicitly wants Meta/Facebook Ad Library results.

1. Start with search_lib_ads or search_adlibrary_products
2. Filter as tightly as possible
3. Use get_lib_ad_detail or get_adlibrary_product_detail for specific items

6. Monitoring Workflow

Use when the user wants ongoing tracking instead of one-time research.

1. Identify the Facebook advertiser with search_fb_advertisers
2. Create a task using create_monitor_task
3. Organize tasks with groups if needed
4. Review ongoing results via monitor overview, ad list, product list, landing page list, or deep analysis tools
5. Cancel tasks with cancel_monitor_task when the user no longer needs tracking

7. AI Image Search Workflow

Use when the user provides an image and wants visually similar ads, products, stores, or TikTok Shop items.

1. Submit the image with ai_search_submit_image
2. Check processing state with ai_search_image_status
3. Read the overview with ai_search_image_summary
4. Fetch the matching result type:

- ads → ai_search_image_ads - products → ai_search_image_products - stores → ai_search_image_stores - TikTok products → ai_search_image_tiktok_products - TikTok shops → ai_search_image_tiktok_shops

Monitor Task Notes

The following tools modify remote state and should be used intentionally:

• create_monitor_task
• cancel_monitor_task
• set_monitor_task_group
• create_monitor_group
• update_monitor_group
• delete_monitor_group
• save_monitor_notifications

Before creating, updating, grouping, or deleting monitoring resources, make sure the advertiser, group, or notification target is clear.

Usage Examples

• "Search for trending TikTok ads about phone cases in the US"
• "Find Shopify products with over 100k views in the last 30 days"
• "Show me the top advertisers on Facebook this week"
• "Get store competition analysis for store ID c2d5b2547218a"
• "Create a monitor task for advertiser XYZ"
• "Search TikTok Shop products with rising sales in beauty category"

Scope

This skill ONLY:

• Calls the PipiAds API through the configured MCP server
• Searches and analyzes advertising intelligence data available through PipiAds
• Retrieves ads, products, stores, advertisers, apps, rankings, monitor data, and related analysis results exposed by PipiAds
• Sends image inputs only when the user explicitly requests image-based search
• Creates or manages monitoring resources only when the user explicitly asks for that outcome

This skill NEVER:

• Guarantees a product, ad, or store will be profitable
• Replaces general market research outside supported PipiAds data
• Scrape websites directly outside the PipiAds service
• Access local files unless the user explicitly provides an image for upload
• Send credentials other than the configured PIPIADS_API_KEY
• Assume broad default search scopes when key constraints are missing
• Use monitor-management tools unless the requested change is clear

Credits

Each API call consumes credits from your PipiAds account:

• List/Search: 1 credit per result returned
• Detail: 1 credit per request (free if queried within 3 days)