Adology AI
Perplexity for social.
От сообщества: Добавлен пользователем или импортирован; проверьте владельца перед подключениемРаботаетБез входаГлобальныйБесплатноМожет изменять данные
Что умеет
- Whoami: Returns information about the currently authenticated user including their team, name, email, and credit balance. Use this to verify connection and understand who you are acting on behalf of.
- List Knowledge Sets: Lists knowledge sets owned by the authenticated user. Returns name, description, feed counts, and configHealth status (needs_feeds, needs_sources, or ready). Start here to see wha
- Get Knowledge Set: Gets full details of a knowledge set including all feeds (brands, influencers, search terms, discussions) with their sources and per-feed item statistics. Use this to understand wha
Какие данные видит
Нужен ли аккаунт
Не нужен: сервер работает без входа
Perplexity for social. Ask anything about what's happening across social and get answers grounded in real ad and creative data, not guesses.
Adology watches the brands, creators, and topics you care about across TikTok, Instagram, Meta Ads, YouTube, and more, then answers your questions the way you'd ask them.
Use it when you need to:
- See what a competitor is running: their current and past ads, the hooks and angles they lean on, and how their messaging is shifting.
- Get the competitor breakdown done: the quarterly or monthly recap nobody has time to build, on demand.
- Build the Monday recap: what moved last week, ready to drop into your inbox or a client's.
- Filter by your own rules: scan all of social and screen out the noise, like boosted posts that pollute your analytics, using your own metrics instead of canned dashboards.
- Break down why creative works: the format, hook, pacing, and CTA behind any piece of content.
- See into the walled gardens: what's actually happening around any brand across platforms.
- Find who's worth watching: discover brands and creators to track, then pull performance from your own ad accounts.
Requires a free Adology account at https://dash.adologyai.com/explorer?invite_code=EXPLORE2026? Authenticate via OAuth when you connect.
Список инструментов сервера (32)
Технические названия из tools/list. Нужны только разработчикам.
| whoami | Returns information about the currently authenticated user including their team, name, email, and credit balance. Use this to verify connection and understand who you are acting on behalf of. |
| list_knowledge_sets | Lists knowledge sets owned by the authenticated user. Returns name, description, feed counts, and configHealth status (needs_feeds, needs_sources, or ready). Start here to see what data is available before running analysis. Use get_knowledge_set for detailed per-feed audit of a specific knowledge set. |
| get_knowledge_set | Gets full details of a knowledge set including all feeds (brands, influencers, search terms, discussions) with their sources and per-feed item statistics. Use this to understand what data is available before running analysis. |
| create_knowledge_set | Creates a new empty knowledge set. After creating, use add_feed to add brands, influencers, search terms, or discussion feeds. |
| add_feed | Adds a feed to a knowledge set. Feed types: "brand" (company ad library + organic social), "influencer" (creator content), "search" (keyword monitoring across platforms), "discussion" (Reddit subreddit). For brand/influencer feeds, provide the name and platform handle/ID. For search feeds, provide the search term. For discussion feeds, provide the subreddit name. |
| remove_feed | Removes a feed from a knowledge set by name and type. This does not delete the underlying data — it just stops tracking that feed in this knowledge set. |
| batch_add_feeds | Add multiple feeds to a knowledge set in a single call. Accepts an array of feed definitions — each with a feedType, name, and platform handles. Much faster than calling add_feed repeatedly. Reads the KS once, applies all feeds, writes once. |
| analyze | Primary tool for content intelligence. Returns posts with AI-generated creative analysis (hookMechanism, creativeConcept, adDescription, transcript, oneLineInsight, etc). Three modes: (1) sampled (default) — control sampling with distribution: "balanced" for equal per-feed representation, "top" for highest engagement per feed (use sortMetric to change which metric), "recent" for newest content; (2) semantic — set mode="semantic" to find posts by natural-language query (keyword fallback when embeddings are unavailable); (3) by-id — pass itemIds to fetch specific items as a deep dive (full creative, labels, performance). In sampled mode all filters (feedNames, platform, date, labelFilter, mediaTypeFilter, outlierFilter) narrow BEFORE sampling. Use outlierFilter to keep only viral items above an engagement-multiple threshold, and labelFilter {filters,mode} for multi-dimension AND/OR label criteria. (semantic mode honors only platform + date + field selection; by-id ignores filters.) Compose filters + distribution for precise queries: e.g. distribution="top" + sortMetric="views" + feedNames=["Celsius"] + platformFilter=["tiktok"]. For label statistics, use get_table_data. |
| get_table_data | Build pivot tables from content data. Like a spreadsheet pivot table — specify rows (what to analyze), columns (how to slice), and metrics (what to measure). Runs server-side in PostgreSQL for fast aggregation. Use listDimensions=true to discover available row dimensions. Use analyze to see actual post content — this tool is for quantifying patterns. |
| compare_knowledge_sets | Compare 2-4 knowledge sets side by side. Returns per-KS summaries including item counts, feed counts, platform breakdown, label coverage, top brands, and brand label profiles. Use this to contrast content strategies, data completeness, or performance across different competitive landscapes. |
| list_labels | List all label dimensions and their top values for a knowledge set. Much lighter than analyze — use this to discover what labels exist before filtering with labelFilter on analyze or get_table_data. |
| search_all | Search for items across ALL your knowledge sets at once. Uses keyword matching to find content matching your query across every KS you own. Results are grouped by knowledge set. Use this when you want to find examples across different competitive landscapes. Note: searches up to 20 knowledge sets, scanning up to 300 recent items per KS. |
| get_creative_dna | Multivariate + combinatorial label analysis: what structural elements drive performance on this KS. Returns a DNA LABEL EFFECTS table joining marginal "average performance" with controlled lift (ridge regression coefficients, independent effects after holding every other label constant), a VS CATEGORY section showing per-brand unique advantages and gaps against the KS baseline, a TRAJECTORY section showing emerging/fading/accelerating labels when a date range is provided, an OPPORTUNITY LABELS section with prioritized ↑LEAN IN / ↓CUT BACK moves, auto-expanded or locked-grid label combinations, and EVIDENCE EXAMPLES pairing top labels with specific post ids. Every cell carries bootstrap confidence; model R² calibrates the whole output. Use focusCategories to lock the analysis to specific axes: ["Hook","Emotion","Production"] builds a 3-way grid. Filter by brand/influencer/search/discussion via feedTypes and feedNames. |
| discover_brands | Search for brands by keyword. Searches the brand discovery database, falling back to brands already tracked in existing knowledge sets. Returns matching brands with their social platform handles/IDs ready for add_feed. Pass brandId to add_feed to auto-resolve all handles, or use the individual handle fields directly. |
| discover_creators | Search for creator profiles across Instagram, LinkedIn, and TikTok. Returns profile details including follower counts and verification status. Use this to find influencers to track. |
| trigger_fetch | Trigger data collection (fetching and processing) for a knowledge set. Charges credits based on the feeds being refreshed. When feedNames is provided, only those specific feeds are refreshed (default and recommended). When omitted, refreshes ALL feeds. Use get_workflow_status to check progress. |
| get_workflow_status | Check the status of an async workflow (export, fetch, label, etc.). Returns the current status and any available progress information. |
| save_to_collection | Save content items to a named collection within a knowledge set. Creates the collection if it does not exist, then adds the specified items. Collections help organize and curate content for later review. |
| get_suggestions | Get data-driven suggestions for enhancing a knowledge set. Returns recommended brands to add, search terms to monitor, influencers to track, and data quality warnings (brands missing sources, missing feed types). Requires the knowledge set to have at least one brand feed configured. |
| list_collections | List all saved collections for a knowledge set. Returns collection names, item counts, and timestamps. Use get_collection to see items in a specific collection. |
| get_collection | Get all items in a saved collection. Returns the full item list with metadata. Use list_collections first to find collection IDs. |
| list_workflows | List all recent workflows (fetch, export, label) for the authenticated user. Shows status (Running, Completed, Failed, etc.), start/close times, and workflow IDs. Use this to check on all pending operations at once instead of calling get_workflow_status for each one. |
| list_conversations | List past chatbot analysis conversations for the authenticated user. Optionally filter by knowledge set. Returns conversation summaries with message counts, sorted by most recently updated. |
| get_conversation | Load the full content of a specific chatbot conversation including all messages and any evidence items attached when the conversation was shared. Use list_conversations first to find conversation IDs. |
| content_intelligence_search | Cross-database content discovery using AI-powered aspect decomposition. Breaks queries into semantic dimensions and searches across vector embeddings + full-text. Use channel param to focus on paid ads, organic posts, search results, or discussions. Returns thumbnails and URLs for rendering. Use analyze with mode:"semantic" for KS-scoped search. |
| list_ad_accounts | List the team's connected Facebook Ads accounts with current sync status. Use as the entry point to discover which ad accounts are queryable. Each account reports lastSyncedAt (when data was last synced into Adology) and dataThroughDate (the latest day of performance data available) — use dataThroughDate to judge how recent an account’s data is. |
| get_account_summary | Returns a one-row rollup of paid performance over a date range: spend, delivery (impressions/clicks/ctr/cpc/cpm), active ad count, and purchase totals + ROAS for the requested attribution window. Use to answer "how are my ads doing overall?" before drilling into individual ads. |
| get_ad_performance | Top-N per-(ad, day) performance rows over a date range, sorted by spend, impressions, or ctr. Returns the standard delivery columns (impressions, clicks, spend, ctr, cpc, cpm, reach, frequency). Use for "which ads spent the most last week" or "highest CTR ads in the period". |
| get_ad_detail | Returns a single ad with daily spend time series + per-event conversion rollup + the latest creative analysis. Composes getAd, getCreative, getSpendTimeSeries, and getConversionsByEvent into one round-trip. Use after list_ad_accounts / get_ad_performance identifies an ad worth drilling into. |
| get_conversions | Per-(ad, day, event) conversion rows for a date range and attribution window. Optional event / ad allow-lists narrow the result. Returns the flexible primitive that other paid tools compose — use for ROAS math, purchase-only filtering, or ad-level conversion drilldowns. |
| get_conversion_funnel | Ordered funnel over a list of canonical events (e.g. view_content → add_to_cart → initiate_checkout → purchase). Returns per-step count, value, and conversion rate vs the prior step (null on the first step or when the prior step is 0). Use to answer "where are users dropping off". |
| list_conversion_events | Lists which canonical conversion events have data for the team over a date range, with availability flags for count, value, and cost-per-event. Omit attributionWindow for broad cross-window discovery, or pass the window you will query next to get window-accurate flags (e.g. the synthetic default window never reports cost-per-event). Use as a discovery step before get_conversions / get_conversion_funnel. |