@appfigures/cli

The Appfigures CLI — query app metrics, reviews, and store data from your terminal.

Try it now without installing:

npx @appfigures/cli auth login
npx @appfigures/cli apps search "youtube"

Install

npm install -g @appfigures/cli

Requires Node.js 22+. Works with pnpm and yarn too.

Quick start

appfigures auth login
appfigures --help

Also available as af alias.

Authentication

• Log in yourself. af auth login --interactive opens your browser in a guided flow; approve and paste the code back.
• Or let your agent guide you. By default, af auth login prints an authorization URL — open it, approve, and finish with af auth login --code <code> using the code shown. An agent can drive this end to end. For unattended use, set APPFIGURES_API_KEY instead: create a token at appfigures.com/developers/keys by clicking Create a New Client, then Create Personal Access Token.

Either login saves the token to your OS credential manager (macOS Keychain, Windows Credential Manager, Linux Secret Service). See Environment for APPFIGURES_API_KEY and other overrides.

MCP server

af mcp runs a local Model Context Protocol server over stdio, exposing the CLI's app-intelligence commands as MCP tools. Point any MCP client (Claude Desktop, Cursor, and others) at it to let an agent query app metrics, reviews, and store data directly.

Add it to your client's MCP config:

{
	"mcpServers": {
		"appfigures": {
			"command": "npx",
			"args": ["-y", "@appfigures/cli", "mcp"]
		}
	}
}

The server signs in with your stored credentials, so run af auth login once first. For a headless setup, pass a token through the client's env instead:

{
	"mcpServers": {
		"appfigures": {
			"command": "npx",
			"args": ["-y", "@appfigures/cli", "mcp"],
			"env": { "APPFIGURES_API_KEY": "<your-token>" }
		}
	}
}

Installed the CLI globally instead of running it through npx? Use "command": "af" with "args": ["mcp"].

Commands

Apps

Find apps and look up their identity. Other commands take the app IDs these return.

Command	Description
af apps search	Find apps by name or publisher. Returns one row per unified app. Default returns Apple and Google listings; pass --all-stores to include other storefronts. To filter apps by estimate values (e.g. apps with >100k downloads last month) use explorer list-products. For estimates broken down by time, country, or storefront, use metrics query with datasets estimates.sales or estimates.revenue.
af apps tracked	List the apps your Appfigures account tracks.
af apps get	Get an app's record: basic metadata (name, developer, etc) and, if the user tracks it, what data they can access. Pass a product ID for one storefront; unified app ID for all storefronts together.

Explorer

Search and analyze the full app catalog: millions of products across Apple, Google Play, Amazon, and other major stores, with 120+ fields spanning identity, storefront and country availability, categories, ratings, release dates, chart ranks, download and revenue estimates, SDK presence, demographics, and related apps.

Command	Description
af explorer list‑products	Read catalog fields for one app or many. Fields referenced by query or sort come back automatically; pass --extra-fields for more. Use ["match","product_id",<id>] for a single app, or combine filters for population queries (e.g. iOS apps using Firebase with $1M+ US revenue). The 120+ fields span ranks, ratings, download and revenue estimates, SDKs, demographics, and more; query grammar and field list in docs get catalog_playbook.
af explorer aggregate‑products	Aggregate across the full catalog of millions of products across Apple, Google Play, Amazon, and other major stores: counts, averages, min/max, and histograms over any set of matching products. Uses the same query grammar as explorer list-products; returns aggregates, not product records. For market sizing, benchmarking, and segment analysis.
af explorer describe‑fields	List the catalog fields and the current user's access level for each. Same field set explorer list-products and explorer aggregate-products accept. Pass --q to search the catalog by keyword.

Metrics

Query numeric datasets across dimensions.

Command	Description
af metrics query	Query any numeric dataset for one or more apps. Optionally grouped by up to two dimensions, returned as a nested partition tree, not app records. Independently filterable by country, device type, and date range. filterAppsBy* options narrow the app set (by ID, storefront, source, or type). Some datasets require app ownership; others work for any app (with the right plan).

Store

App store presence: listing content, category ranks, top charts, and featured placements.

Command	Description
af store app‑ranks	Trace rank history for one or more apps across countries, device types, category subtypes, and categories, as time-series positions with day-over-day deltas.
af store top‑charts	List the top apps in a category chart for a given country and category, with current positions and day-over-day deltas.
af store categories	List every store category with its ID. Numeric category IDs required by store app-ranks --category-ids and store top-charts --category-id are available here.
af store featured	List featured and editorial placements for an app or storefront product. Pass --count=0 for summary stats only.
af store app‑listing	Read the full store listing for one storefront: localized text (name, subtitle, description, release notes) plus screenshots, video, categories, monetization, supported devices, country availability, price, file size, and age rating. Takes a numeric product ID (one storefront at a time; a unified app has one product per storefront). One locale per request.

Reviews

Search, summarize, and reply to iOS and Google Play app store reviews.

Command	Description
af reviews list	Read individual reviews for one or more apps. Returns review text, star rating, country, and app version. Filterable by star rating, date range, country, version, and tracking relationship.
af reviews breakdown	Aggregate review counts for one or more apps, bucketed by dimension. Returns one count per dimension value, plus a global total across the matched set.
af reviews reply	Post or withdraw a developer response on a specific review. Pass content to post; pass delete: true to withdraw a previously-posted response. Returns the resulting state (published/pending for a post, removed/removal_pending for a withdrawal) along with the submitting account.

Keywords

Organic keyword visibility and tracking.

Command	Description
af keywords list	List tracked keywords with their opaque IDs.
af keywords rankings	Check the organic keywords one or more apps rank for, with position, popularity, and competitiveness.

Sdks

Look up the SDKs we track.

Command	Description
af sdks list	List every known SDK with its id, or search to find a specific one.

Docs

Reference docs and guides for specific actions and common tasks.

Command	Description
af docs get	Return a reference doc or guide by slug.

API

Command	Description
af api	Make a raw API request for endpoints without a dedicated command. Endpoints, parameters, and response shapes are documented at https://docs.appfigures.com.

MCP

Command	Description
af mcp	Run an MCP server over stdio for MCP clients like Claude Desktop and Cursor to call Appfigures tools.

Auth

Command	Description
af auth login	Sign in to Appfigures
af auth logout	Remove stored credentials
af auth status	Show authentication and account status

Run af <command> --help for arguments, flags, and examples.

Environment

Variable	Purpose
APPFIGURES_API_KEY	API key; skips interactive auth
AF_VERBOSE	Log HTTP to stderr (same as -v)
NO_COLOR	Disable ANSI color
NO_UPDATE_NOTIFIER	Skip the npm-registry update check
CI	Also skips the update check (any CI system)

API Reference

Every command with its full argument and flag list. For the one-line overview, see Commands above.

Global flags. All commands accept:

• -v, --verbose — Log HTTP requests to stderr. Also set via AF_VERBOSE=1.
• -V, --version — Print the CLI version and exit.
• -h, --help — Show usage for the current command.

Output format. Every command emits a single JSON value on stdout — pipe to jq for filtering. Informational messages, hints, and update notices go to stderr so pipelines stay clean.

af apps search

af apps search <query> [flags]

Find apps by name or publisher. Returns one row per unified app. Default returns Apple and Google listings; pass --all-stores to include other storefronts. To filter apps by estimate values (e.g. apps with >100k downloads last month) use explorer list-products. For estimates broken down by time, country, or storefront, use metrics query with datasets estimates.sales or estimates.revenue.

Options

• <query> required string. Search query (app name or publisher).
• --all-stores boolean, default false. Include storefronts beyond Apple and Google: Amazon, Windows, Steam, Roku, LG TV, Samsung TV, and others.
• --count integer, default 10. Number of results to return
• --page integer, default 1. Page number.

Examples

# Find every Electronic Arts app.
af apps search 'electronic arts'

# Page through long results.
af apps search 'electronic arts' --count=25 --page=2

# Find Minecraft on every storefront (e.g. Amazon, Steam, Windows, Roku; not common).
af apps search minecraft --all-stores

---

af apps tracked

af apps tracked [flags]

List the apps your Appfigures account tracks.

Options

• --count integer, default 10. Number of results to return
• --page integer, default 1. Page number.
• --q string. App name to filter by.
• --filter-apps-by-id (integer or string)[]. Only include data about specific apps, by product ID or unified app ID. Takes precedence over the other filterAppsBy* keys when set. Storefront, source, or type filters are better for app sets that can be described by those criteria.
• --filter-apps-by-storefront string[]. Narrow the account's tracked apps to those on these storefronts (e.g. apple:ios, google_play).
• --filter-apps-by-source string[]. Narrow the account's tracked apps by tracking relationship.
• --filter-apps-by-type string[]. Narrow the account's tracked apps to products of these types.

Examples

# List your apps with private-data access.
af apps tracked --filter-apps-by-source=own,shared

# List your tracked fitness apps.
af apps tracked --q=fitness

# List just your iOS apps.
af apps tracked --filter-apps-by-storefront=apple:ios

# Page through long results.
af apps tracked --filter-apps-by-source=own,shared --count=50 --page=2

# List tracked competitors.
af apps tracked --filter-apps-by-source=manual

# Find individual IAPs or subscriptions (not common).
af apps tracked --filter-apps-by-type=inapp,subscription

---

af apps get

af apps get <app-id> [flags]

Get an app's record: basic metadata (name, developer, etc) and, if the user tracks it, what data they can access. Pass a product ID for one storefront; unified app ID for all storefronts together.

Options

• <app-id> required integer or string. App identifier. Unified app ID or product ID.
• --all-stores boolean, default false. For a unified app ID: include member products across all storefronts (Amazon, Steam, Windows, Roku, etc.). When false, member_products is restricted to storefronts with app-intelligence coverage (iOS + Google Play). Ignored for product IDs.

Examples

# Get Minecraft's unified-app record (iOS + Google Play by default).
af apps get ua_X7iNgb

# Get Minecraft's Google Play product record.
af apps get 6938219

---

af explorer list-products

af explorer list-products [flags]

Read catalog fields for one app or many. Fields referenced by query or sort come back automatically; pass --extra-fields for more. Use ["match","product_id",<id>] for a single app, or combine filters for population queries (e.g. iOS apps using Firebase with $1M+ US revenue). The 120+ fields span ranks, ratings, download and revenue estimates, SDKs, demographics, and more; query grammar and field list in docs get catalog_playbook.

Options

• --query array. Explorer query in JSON array format to select matching catalog Products. Missing values and [] match every Product across every storefront. The full field list and query syntax are documented in docs get catalog_playbook.
• --extra-fields string[]. Additional fields to include. The full field list is documented in docs get catalog_playbook.
• --sort string. Explorer field name. The full field list is documented in docs get catalog_playbook.
• --order string, default desc. Sort direction.
• --count integer, default 10. Number of results to return
• --page integer, default 1. Page number.
• --allow-unscoped-nested boolean, default false. Escape hatch for intentionally broad queries. Bypasses the default block on unscoped nested predicates that usually inflate results.

Examples

# Find iOS apps that have Firebase installed.
af explorer list-products --query='["and",["match","storefronts","apple:ios"],["nested","all_sdks",["and",["match","all_sdks.id","firebase"],["match","all_sdks.active",true]]]]'

# Rank the biggest US iOS games by revenue.
af explorer list-products --query='["and",["match","storefronts","apple:ios"],["match","categories.all",6014]]' --sort='custom_meta[country=us].revenue_estimates_sum_30_days' --order=desc --count=25

# Find US iOS apps in the $100k–$1M/month revenue tier.
af explorer list-products --query='["and",["match","storefronts","apple:ios"],["nested","custom_meta",["and",["match","custom_meta.country","us"],["match","custom_meta.revenue_estimates_sum_30_days",["number_range",100000,1000000]]]]]'

# Page through results.
af explorer list-products --query='["and",["match","storefronts","apple:ios"],["match","categories.all",6014]]' --count=50 --page=2

# Pass `--extra-fields` for columns the query doesn't already reference. Common for single-app reads.
af explorer list-products --query='["match","product_id",304004187384]' --extra-fields='custom_meta[country=zz].revenue_estimates_sum_365_days,all_sdks[id=firebase].active'

---

af explorer aggregate-products

af explorer aggregate-products <fields> [flags]

Aggregate across the full catalog of millions of products across Apple, Google Play, Amazon, and other major stores: counts, averages, min/max, and histograms over any set of matching products. Uses the same query grammar as explorer list-products; returns aggregates, not product records. For market sizing, benchmarking, and segment analysis.

Options

• --query array. Explorer query in JSON array format to select matching catalog Products. Missing values and [] match every Product across every storefront. The full field list and query syntax are documented in docs get catalog_playbook.
• <fields> required string[]. Field+aggregation pairs (e.g. all_rating/stats, storefronts/terms). Aggregations: stats, terms, histogram, date_histogram, cardinality. The full field list is documented in docs get catalog_playbook.
• --allow-unscoped-nested boolean, default false. Escape hatch for intentionally broad queries. Bypasses the default block on unscoped nested predicates that usually inflate results.
• --terms-count integer, default 20. Maximum buckets returned for each terms aggregation. Other aggregation types ignore it.
• --date-histogram-interval string. Bucket granularity for each date_histogram aggregation. Other aggregation types ignore it.

Examples

# How many monthly downloads does an average iOS app get in Japan?
af explorer aggregate-products 'custom_meta[country=jp].download_estimates_average_30_days/stats' --query='["and",["match","storefronts","apple:ios"],["match","countries","jp"]]'

# What's the rating, category mix, and developer concentration for US iOS apps in the $100k–$10M/mo net-revenue tier?
af explorer aggregate-products all_rating/stats,categories.all/terms,developer_id/cardinality --query='["and",["match","storefronts","apple:ios"],["nested","custom_meta",["and",["match","custom_meta.revenue_estimates_sum_30_days",["number_range",100000,10000000]],["match","custom_meta.country","us"]]]]'

# Are new iOS games still launching at the same rate as two years ago?
af explorer aggregate-products release_date/date_histogram --query='["and",["match","storefronts","apple:ios"],["match","categories.all",6014],["match","release_date",["range","2024-01-01","2025-12-31"]]]'

# What SDKs do apps commonly ship alongside OneSignal?
af explorer aggregate-products 'all_sdks[*].id/terms' --query='["nested","all_sdks",["and",["match","all_sdks.id","onesignal"],["match","all_sdks.active",true]]]'

# How do iOS app ratings distribute?
af explorer aggregate-products all_rating/histogram --query='["match","storefronts","apple:ios"]'

# How many apps are on each storefront?
af explorer aggregate-products storefronts/terms

---

af explorer describe-fields

af explorer describe-fields [flags]

List the catalog fields and the current user's access level for each. Same field set explorer list-products and explorer aggregate-products accept. Pass --q to search the catalog by keyword.

Options

• --count integer, default 50. Number of results to return
• --page integer, default 1. Page number.
• --q string. Filter by path, title, description, type.

Examples

# Search for revenue-related fields.
af explorer describe-fields --q=revenue

# List every catalog field with the current user's access level.
af explorer describe-fields

---

af metrics query

af metrics query <dataset> [flags]

Query any numeric dataset for one or more apps. Optionally grouped by up to two dimensions, returned as a nested partition tree, not app records. Independently filterable by country, device type, and date range. filterAppsBy* options narrow the app set (by ID, storefront, source, or type). Some datasets require app ownership; others work for any app (with the right plan).

Options

• <dataset> required string. Dataset to query (e.g. sales.combined_downloads). The full list is documented in docs get numeric_metrics.
• --group-by string[]. Dimensions to group by. Max 2: the first slot becomes the outer entity type, the second the inner series.
• --granularity string. Time granularity when grouping by date
• --count integer. Row cap. With --group-by, top N of the outer entity type by value (earliest N when grouping by date). Without --group-by, single-page preview.
• --countries string[]. Filter to one or more ISO country codes (e.g. US, JP, GB)
• --device-type string. Device type
• --all-time boolean, default false. Opt in to the entire history. Without this flag (and without start/end), the query defaults to the last 30 days. Mutually exclusive with start and end.
• --filter-apps-by-id (integer or string)[]. Only include data about specific apps, by product ID or unified app ID. Takes precedence over the other filterAppsBy* keys when set. Storefront, source, or type filters are better for app sets that can be described by those criteria.
• --filter-apps-by-storefront string[]. Narrow the account's tracked apps to those on these storefronts (e.g. apple:ios, google_play).
• --filter-apps-by-source string[]. Narrow the account's tracked apps by tracking relationship.
• --filter-apps-by-type string[]. Narrow the account's tracked apps to products of these types.
• --start string. Start date (YYYY-MM-DD)
• --end string. End date (YYYY-MM-DD, defaults to today)

Examples

# Get total downloads across your apps with private-data access.
af metrics query sales.combined_downloads --filter-apps-by-source=own,shared

# Get revenue split by storefront, plus a top-level total.
af metrics query sales.combined_revenue --filter-apps-by-source=own,shared --group-by=storefront

# Rank the top 5 tracked competitors by estimated monthly revenue.
af metrics query estimates.revenue --filter-apps-by-source=manual --group-by=product --count=5

# Track Candy Crush Saga's daily download estimates.
af metrics query estimates.sales --filter-apps-by-id=ua_V1Q1uX --group-by=date --granularity=daily

# Track net monthly recurring revenue per app, month over month.
af metrics query subscriptions.mrr --filter-apps-by-source=own,shared --group-by=product,date --granularity=monthly

# Get Minecraft's new ratings.
af metrics query ratings.new_total --filter-apps-by-id=ua_X7iNgb

# Track daily ad spend across your apps.
af metrics query adspend.cost --filter-apps-by-source=own,shared --group-by=date --granularity=daily

# Compare Candy Crush's December 2025 downloads across the US, Japan, and UK.
af metrics query estimates.sales --filter-apps-by-id=ua_V1Q1uX --countries=US,JP,GB --group-by=country --start=2025-12-01 --end=2025-12-31

# Track all-time monthly revenue across your apps with private-data access.
af metrics query sales.combined_revenue --filter-apps-by-source=own,shared --group-by=date --granularity=monthly --all-time

# Get Minecraft's review volume by country.
af metrics query reviews.total --filter-apps-by-id=ua_X7iNgb --group-by=country

---

af store app-ranks

af store app-ranks <app-ids> [flags]

Trace rank history for one or more apps across countries, device types, category subtypes, and categories, as time-series positions with day-over-day deltas.

Options

• <app-ids> required (integer or string)[]. App identifiers (unified app IDs or product IDs)
• --countries string[]. Country codes to query. Defaults to every country with rank coverage.
• --granularity string, default hourly. Sampling rate. Hourly gives the freshest data; pass --granularity=daily for compact multi-day history.
• --device-types string[], default ["handheld"]. Filter response rows by device type.
• --subtypes string[], default ["free"]. Filter response rows by category subtype.
• --category-ids integer[]. Filter response rows by category ID.
• --start string. Start date (YYYY-MM-DD)
• --end string. End date (YYYY-MM-DD, defaults to today)
• --count integer, default 10. Number of results to return
• --page integer, default 1. Page number.

Examples

# Check ChatGPT's current ranks (unified app).
af store app-ranks ua_miTXv6 --countries=US

# Check ChatGPT's current ranks on one storefront.
af store app-ranks 336744124021 --countries=US

# Compare ChatGPT's ranks across the US, UK, and Japan.
af store app-ranks ua_miTXv6 --countries=US,GB,JP

# Check Procreate's paid iPad chart ranks.
af store app-ranks ua_CxA1MS --subtypes=paid --device-types=tablet --countries=US

# Trace ChatGPT's chart history through May 2026.
af store app-ranks ua_miTXv6 --granularity=daily --start=2026-05-01 --end=2026-05-31 --countries=US

# Check ChatGPT's rank in one category (US iOS Productivity).
af store app-ranks 336744124021 --category-ids=6007 --countries=US

---

af store top-charts

af store top-charts [flags]

List the top apps in a category chart for a given country and category, with current positions and day-over-day deltas.

Options

• --country required string. ISO country code (e.g. US, JP, GB)
• --category-id required integer. Store category ID
• --subtype string, default free. Category subtype (chart variant within the category).
• --date string. Snapshot date (YYYY-MM-DD, defaults to current).
• --count integer, default 10. Number of results to return
• --page integer, default 1. Page number.

Examples

# Find the Games category, then pull its US chart.
af store categories --q=games
af store top-charts --country=US --category-id=6014

# List top paid apps on the US App Store.
af store top-charts --country=US --category-id=25204 --subtype=paid

# List top free apps on the Japan App Store.
af store top-charts --country=JP --category-id=25204

# List top free apps on Google Play in the US.
af store top-charts --country=US --category-id=100

# List top free apps on the US App Store in May 2026.
af store top-charts --country=US --category-id=25204 --date=2026-05-01

---

af store categories

af store categories [flags]

List every store category with its ID. Numeric category IDs required by store app-ranks --category-ids and store top-charts --category-id are available here.

Options

• --count integer, default 50. Number of results to return
• --page integer, default 1. Page number.
• --q string. Filter by name.
• --sort-by string. Sort order. Default: relevance when q is set, otherwise list order.
• --id integer[]. Only return these category IDs.
• --parent-id integer. Only include subcategories of this parent category (drill-down by id).
• --storefront string[]. Only include categories from these storefronts (e.g. apple:ios, google_play).
• --device-type string[]. Only include categories for these device types (e.g. handheld, tablet).
• --all boolean, default false. Include non-rank stores (roku, vizio, etc.). These have categories but no rank data.

Examples

# Find the Games category.
af store categories --q=games

# List every Apple iOS category.
af store categories --storefront=apple:ios

# List subcategories of a parent category (here, Apple Games).
af store categories --parent-id=6014

# Include categories from non-rank-supporting stores (e.g. Roku, Vizio) (not common).
af store categories --all

---

af store featured

af store featured <app-id> [flags]

List featured and editorial placements for an app or storefront product. Pass --count=0 for summary stats only.

Options

• <app-id> required integer or string. App identifier. Unified app ID or product ID.
• --countries string[]. Countries to include. Pass multiple countries to compare markets; omit when using --all-countries.
• --all-countries boolean, default false. Include every country by omitting the --countries filter.
• --include-rank-trend boolean, default false. Include per-interval rank_trend for each placement.
• --sort string, default relevance. Sort placements by relevance or date.
• --order string, default desc. Sort direction.
• --start string. Start date (YYYY-MM-DD)
• --end string. End date (YYYY-MM-DD, defaults to today)
• --count integer, default 10. Number of results to return
• --page integer, default 1. Page number.

Examples

# List Minecraft's recent featured placements (unified app).
af store featured ua_X7iNgb

# List Minecraft's recent featured placements on one storefront.
af store featured 10157213

# Compare Minecraft's placement coverage across the US, UK, and Japan.
af store featured ua_X7iNgb --countries=US,GB,JP

# List Minecraft's placements during a specific month (December 2025).
af store featured ua_X7iNgb --start=2025-12-01 --end=2025-12-31

# List Minecraft's placements with rank history.
af store featured ua_X7iNgb --include-rank-trend

# List Minecraft's placements sorted by end date, newest run first.
af store featured ua_X7iNgb --sort=date --order=desc

# Get Minecraft's placement summary only.
af store featured ua_X7iNgb --count=0

---

af store app-listing

af store app-listing <product-id> [flags]

Read the full store listing for one storefront: localized text (name, subtitle, description, release notes) plus screenshots, video, categories, monetization, supported devices, country availability, price, file size, and age rating. Takes a numeric product ID (one storefront at a time; a unified app has one product per storefront). One locale per request.

Options

• <product-id> required integer. Numeric product ID for one storefront. Not a unified app ID. Member product_id values are available from apps get '<unified-app-id>'.
• --language string. Locale (e.g. en, ja, zh-Hans) for name, subtitle, description, release notes, and screenshots. Defaults to en; falls back to the first available locale when the requested one has no metadata. The response echoes the resolved language.
• --device-type string, default handheld. Relevant to Apple apps. Pick handheld for iPhone-specific metadata, tablet for iPad, desktop for Mac, etc.

Examples

# Read Minecraft's store listing.
af store app-listing 10157213

# Read Minecraft's Japanese-localized listing.
af store app-listing 10157213 --language=ja

# Read Minecraft's iPad screenshots.
af store app-listing 10157213 --device-type=tablet

---

af reviews list

af reviews list [flags]

Read individual reviews for one or more apps. Returns review text, star rating, country, and app version. Filterable by star rating, date range, country, version, and tracking relationship.

Options

• --stars number[]. Filter by star rating.
• --versions string[]. Filter by app version. Pass multiple to combine.
• --countries string[]. Filter to one or more ISO country codes (e.g. US, JP, GB).
• --q string. Search review title and body. Pass multiple keywords to match any. Case-insensitive; combines with other filters.
• --sort string. Sort by review date or star rating.
• --order string, default desc. Sort direction.
• --filter-apps-by-id (integer or string)[]. Only include data about specific apps, by product ID or unified app ID. Takes precedence over the other filterAppsBy* keys when set. Storefront, source, or type filters are better for app sets that can be described by those criteria.
• --filter-apps-by-storefront string[]. Narrow the account's tracked apps to those on these storefronts (e.g. apple:ios, google_play).
• --filter-apps-by-source string[]. Narrow the account's tracked apps by tracking relationship.
• --filter-apps-by-type string[]. Narrow the account's tracked apps to products of these types.
• --start string. Start date (YYYY-MM-DD)
• --end string. End date (YYYY-MM-DD, defaults to today)
• --count integer, default 10. Number of results to return
• --page integer, default 1. Page number. 1-500.

Examples

# Read Minecraft's recent reviews.
af reviews list --filter-apps-by-id=ua_X7iNgb

# Read Minecraft's 5-star reviews.
af reviews list --filter-apps-by-id=ua_X7iNgb --stars=5

# Compare Minecraft's reviews across the US, Japan, and the UK.
af reviews list --filter-apps-by-id=ua_X7iNgb --countries=US,JP,GB

# Read Minecraft's December 2025 reviews.
af reviews list --filter-apps-by-id=ua_X7iNgb --start=2025-12-01 --end=2025-12-31

# Page through long results across your own apps.
af reviews list --filter-apps-by-source=own --count=50 --page=2

---

af reviews breakdown

af reviews breakdown [flags]

Aggregate review counts for one or more apps, bucketed by dimension. Returns one count per dimension value, plus a global total across the matched set.

Options

• --stars number[]. Filter by star rating.
• --versions string[]. Filter by app version. Pass multiple to combine.
• --countries string[]. Filter to one or more ISO country codes (e.g. US, JP, GB).
• --q string. Search review title and body. Pass multiple keywords to match any. Case-insensitive; combines with other filters.
• --filter-apps-by-id (integer or string)[]. Only include data about specific apps, by product ID or unified app ID. Takes precedence over the other filterAppsBy* keys when set. Storefront, source, or type filters are better for app sets that can be described by those criteria.
• --filter-apps-by-storefront string[]. Narrow the account's tracked apps to those on these storefronts (e.g. apple:ios, google_play).
• --filter-apps-by-source string[]. Narrow the account's tracked apps by tracking relationship.
• --filter-apps-by-type string[]. Narrow the account's tracked apps to products of these types.
• --start string. Start date (YYYY-MM-DD)
• --end string. End date (YYYY-MM-DD, defaults to today)
• --by string[]. Limit the response to these dimensions; omit to return all.
• --top integer, default 20. Maximum values returned per dimension; the rest are summed under __other__.

Examples

# Break down Minecraft's recent reviews.
af reviews breakdown --filter-apps-by-id=ua_X7iNgb

# Where are Minecraft's biggest fans writing from?
af reviews breakdown --filter-apps-by-id=ua_X7iNgb --stars=5

# Count Minecraft's December 2025 5-star reviews.
af reviews breakdown --filter-apps-by-id=ua_X7iNgb --stars=5 --start=2025-12-01 --end=2025-12-31

# How many Minecraft reviewers raved?
af reviews breakdown --filter-apps-by-id=ua_X7iNgb --q='love amazing fun great best'

# Compare review volume across your own apps.
af reviews breakdown --filter-apps-by-source=own

---

af reviews reply

af reviews reply <review-id> [flags]

Post or withdraw a developer response on a specific review. Pass content to post; pass delete: true to withdraw a previously-posted response. Returns the resulting state (published/pending for a post, removed/removal_pending for a withdrawal) along with the submitting account.

Options

• <review-id> required string. Review to act on. Use review_id from reviews list.
• --content string. Response text the developer wants to publish.
• --delete boolean. Withdraw the previously-posted response on this review. Mutually exclusive with content.

Examples

# Reply to a low-star review after shipping a fix.
af reviews reply rev123 --content='We just shipped a fix in v2.1. Let us know if you still see this.'

# Withdraw a previously-posted response.
af reviews reply rev123 --delete

---

af keywords list

af keywords list [flags]

List tracked keywords with their opaque IDs.

Options

• --count integer, default 10. Number of results to return
• --page integer, default 1. Page number.
• --q string. Filter by keyword_term.
• --sort-by string. Sort order. Default: relevance when q is set, otherwise list order.
• --include-relationships boolean, default false. Include per-(product, country) tracking detail and sync state on each row. Off by default; adds a nested block per tracked (product, country) pair.

Examples

# List every tracked keyword.
af keywords list

# Search tracked keywords for "fitness".
af keywords list --q=fitness

# List the most-recently-tracked keywords first.
af keywords list --sort-by=-added_on

# Show each keyword's tracking and sync detail.
af keywords list --include-relationships

---

af keywords rankings

af keywords rankings [flags]

Check the organic keywords one or more apps rank for, with position, popularity, and competitiveness.

Options

• --product-ids integer[]. Product identifiers (numeric, one storefront each).
• --country required string. ISO country code (e.g. US, JP, GB)
• --device-type string. Device type
• --count integer, default 10. Number of results to return (min 10).
• --page integer, default 1. Page number.

Examples

# Check ChatGPT's current US keyword rankings.
af keywords rankings --product-ids=336744124021 --country=US

# Check ChatGPT's iPad-only keyword rankings.
af keywords rankings --product-ids=336744124021 --country=US --device-type=tablet

# Compare ChatGPT and Gemini's US keyword rankings.
af keywords rankings --product-ids=336744124021,337217072531 --country=US

---

af sdks list

af sdks list [flags]

List every known SDK with its id, or search to find a specific one.

Options

• --count integer, default 50. Number of results to return
• --page integer, default 1. Page number.
• --q string. Filter by name, description, tags.
• --sort-by string. Sort order. Default: relevance when q is set, otherwise list order.
• --id string[]. Only return these SDK ids.
• --include-inactive boolean, default false. Include inactive SDKs. Rare; most callers want active only.

Examples

# Find OneSignal's id.
af sdks list --q=OneSignal

# Search for analytics SDKs.
af sdks list --q=analytics

# Look up details for several SDK ids.
af sdks list --id=firebase,admob,onesignal

# Include inactive SDKs in the listing (not common).
af sdks list --include-inactive

---

af docs get

af docs get <slug>

Return a reference doc or guide by slug.

Options

• <slug> required string. Which reference to return

Examples

# List every dataset [`metrics query`](#command-metrics-query) accepts.
af docs get numeric_metrics

# Read the catalog query grammar, field-reference syntax, aggregation rules, and worked queries for [`explorer list-products`](#command-explorer-list-products) and [`explorer aggregate-products`](#command-explorer-aggregate-products).
af docs get catalog_playbook

# Read the glossary of domain terms seen in responses.
af docs get glossary

---

af api

af api <path> [flags]

Make a raw API request for endpoints without a dedicated command. Endpoints, parameters, and response shapes are documented at https://docs.appfigures.com.

Options

• <path> required string. API path (e.g. /users, /products)
• --method string, default GET. HTTP method (one of: GET, POST, PUT, PATCH, DELETE, HEAD, OPTIONS)
• --body string. Request body (JSON string)

---

af mcp

af mcp

Run an MCP server over stdio for MCP clients like Claude Desktop and Cursor to call Appfigures tools.

---

af auth login

af auth login [flags]

Sign in to Appfigures

Prints a URL to authorize this CLI. Open it, approve access, then re-run with --code <code> using the code shown after approval. Pass --interactive to sign in through a guided browser flow instead.

For unattended use (CI, scripts), set APPFIGURES_API_KEY in the environment instead. No browser flow needed.

Options

• --code string. Authorization code from the OAuth consent page. Second step of af auth login: exchanges the code, saves the token, exits.
• --interactive boolean. Sign in through a guided browser flow.

---

af auth logout

af auth logout

Remove stored credentials

---

af auth status

af auth status

Show authentication and account status

Support

• Issues: github.com/appfigures/cli/issues
• API docs: docs.appfigures.com

Contributing

This package is developed in a private monorepo and published here as a mirror. To report bugs or request features, open an issue.

We're also hiring AI-native devs. If you want to help build the future of AI App Intelligence apply at appfigures.com/careers.

License

Apache 2.0