@LC_builder

Build rich applications with real-time social data. Powered by @LunarCrush.

 California, United States
LunarCrush.com
Joined July 2019
  • Tweets13.1K
  • Following4
  • Followers15.7K
2.6KMedia
Nobody asks you to add a social data platform. They ask you to make the Monday numbers right. Three drop-in points: the MCP server for the assistant your team uses, the CLI in cron for the warehouse load, the REST API for the app. One key covers all three.
Flow diagram of four drop-in points: the MCP server for Claude, ChatGPT and Gemini, the CLI in cron, the REST API v4, and output to Slack and Zapier, all on one key.

ALT Flow diagram of four drop-in points: the MCP server for Claude, ChatGPT and Gemini, the CLI in cron, the REST API v4, and output to Slack and Zapier, all on one key.

1
1
116
Claude, ChatGPT, Gemini, Slack and Zapier all connect from this page. lunarcrush.com/en/integratio…

Connect LunarCrush to your stack

Claude, ChatGPT, Gemini, Slack, Zapier, and beyond — via MCP and API.

lunarcrush.com
17
No client library, no auth plumbing, no parser. lunarcrush creators ferrari --limit 10 --format json 15,570 accounts mentioned Ferrari in the last 24 hours as of this morning. That returns the ten that mattered, as JSON, straight into jq. Same key as the API and MCP server.
Terminal card showing the lunarcrush creators ferrari command returning ten rows out of 15,570 creators in 24 hours, with the top account at 46,900 followers and 6.54M interactions.

ALT Terminal card showing the lunarcrush creators ferrari command returning ten rows out of 15,570 creators in 24 hours, with the top account at 46,900 followers and 6.54M interactions.

1
116
Install is one line on macOS and Linux, and it ships with every paid plan. lunarcrush.com/en/developers…

Build with LunarCrush: API, MCP, and SDKs

Social sentiment, market data, creators, and trending topics for crypto and stocks — one API. MCP-ready for Claude, ChatGPT, and Gemini.

lunarcrush.com
16
Mapping a sector usually means a research vendor and six weeks. /category/ai/topics/v1 returns the shape in one call. OpenAI at 66,095 creators in 24 hours, Anthropic at 45,208, then a tail that runs down to single digits. Forty other categories, same call.
Table of the AI category by creators in 24 hours: OpenAI 66,095 and 204.8M interactions, Anthropic 45,208 and 194.1M, Reddit Inc 27,869, DeepSeek 9,338, Nebius Group 3,378.

ALT Table of the AI category by creators in 24 hours: OpenAI 66,095 and 204.8M interactions, Anthropic 45,208 and 194.1M, Reddit Inc 27,869, DeepSeek 9,338, Nebius Group 3,378.

128
Follower count is a weak filter for influence. Rolex, 24h as of this morning. The top-ranked creator has 2,474 followers and 2.61M engagements off 9 posts. Second place has 1.02M followers and 2.34M off 4. creator_rank scores the posts. creator_followers scores the audience.
Scatter plot of the top 10 Rolex creators over 24 hours, with the rank 1 account at 2,474 followers and 2.61M interactions against rank 2 at 1.02M followers and 2.34M.

ALT Scatter plot of the top 10 Rolex creators over 24 hours, with the rank 1 account at 2,474 followers and 2.61M interactions against rank 2 at 1.02M followers and 2.34M.

175
One account can be the entire number. Hasselblad, last 24 hours as of this morning: 29.7M engagements, 832 creators. A single TikTok account drove 83.8% of it off 17 posts. Twelve lines and two endpoints catch that before the readout does.
Funnel showing the Hasselblad topic at 29.7M engagements from 832 creators in 24 hours, with one account posting 17 times and driving 83.8 percent of the total.

ALT Funnel showing the Hasselblad topic at 29.7M engagements from 832 creators in 24 hours, with one account posting 17 times and driving 83.8 percent of the total.

1
1
166
Swap the topic string for whatever you monitor. Anything over about 50% is one account, not a trend.
Twelve lines of Python that call the topic and topic creators endpoints and return the top creator's share of 24 hour interactions, printing 83.8 for hasselblad.

ALT Twelve lines of Python that call the topic and topic creators endpoints and return the top creator's share of 24 hour interactions, printing 83.8 for hasselblad.

13
One tool call tells you where a conversation actually lives. Prompt your MCP client: break DeepSeek down by network, last 24 hours. Tool: topic, backed by /topic/deepseek/v1. X: 84% of mentions, 73% of engagements. YouTube: 8% of mentions, 20% of engagements.
Bar chart of DeepSeek interactions by network over 24 hours: X 11M at 73 percent, YouTube 3.1M at 20 percent, TikTok 1.1M at 7 percent, Instagram 17K, Reddit 16K, news 920.

ALT Bar chart of DeepSeek interactions by network over 24 hours: X 11M at 73 percent, YouTube 3.1M at 20 percent, TikTok 1.1M at 7 percent, Instagram 17K, Reddit 16K, news 920.

1
157
Setup for Claude Desktop and Cursor, plus any client that takes a URL. lunarcrush.com/en/developers…

Plug LunarCrush into any MCP client

Tool definitions, auth, and example workflows for Claude, ChatGPT, Gemini, and any MCP-compatible client.

lunarcrush.com
12
An alert that fires on hour-old data is not an alert. /coins/list/v1 is cached up to an hour. /coins/list/v2 refreshes every few seconds. Bitcoin, 24h as of this morning: 133.7M engagements, 49,496 creators. v1 for dashboards. v2 for triggers.
Table comparing coins list v1, cached up to one hour, with v2, refreshing every few seconds, noting Bitcoin at 133.7M engagements and 49,496 creators in 24 hours.

ALT Table comparing coins list v1, cached up to one hour, with v2, refreshing every few seconds, noting Bitcoin at 133.7M engagements and 49,496 creators in 24 hours.

1
1
300
The v2 page lists the sort and filter params that make it usable. lunarcrush.com/en/developers…

Build with LunarCrush: API, MCP, and SDKs

Social sentiment, market data, creators, and trending topics for crypto and stocks — one API. MCP-ready for Claude, ChatGPT, and Gemini.

lunarcrush.com
7
Two brands, near identical engagement totals, completely different conversations. /category/technology-brands/topics/v1 this morning: Adobe, 84.4M engagements from 5,342 creators. Anthropic, 81.2M from 36,676 creators. 7x apart on breadth. Chart creators beside the total.
Table comparing Adobe at 84.4M engagements from 8,037 mentions and 5,342 creators with Anthropic at 81.2M engagements from 63,295 mentions and 36,676 creators.

ALT Table comparing Adobe at 84.4M engagements from 8,037 mentions and 5,342 creators with Anthropic at 81.2M engagements from 63,295 mentions and 36,676 creators.

146
A VP asks which of our names people are unhappy with. That used to be a ticket for the data team. In Claude, the MCP stocks tool calls /stocks/list/v2. Of the 15 most discussed names this morning: Reddit 48% sentiment, Uber 51%, Honda 56%, PepsiCo 90%. One prompt.
An MCP conversation in Claude where the stocks tool returns Reddit at 48% sentiment, Uber at 51%, Honda at 56% and PepsiCo at 90% across the 15 most discussed stocks.

ALT An MCP conversation in Claude where the stocks tool returns Reddit at 48% sentiment, Uber at 51%, Honda at 56% and PepsiCo at 90% across the 15 most discussed stocks.

1
160
The MCP server runs on the same key as the API and works in any client that accepts a URL. lunarcrush.com/en/developers…

Plug LunarCrush into any MCP client

Tool definitions, auth, and example workflows for Claude, ChatGPT, Gemini, and any MCP-compatible client.

lunarcrush.com
16
The slug you guess is not always the entity you want. /topic/openai/v1 returned 17 mentions in 24 hours. /topic/open-ai/v1 returned 95,475. One is a conference topic. One is the company. Both are valid. One is your dashboard. Resolve slugs at setup.
Table comparing two requests: /topic/openai/v1 returning OpenAI Dev Day with 17 mentions and 14 creators, and /topic/open-ai/v1 returning OpenAI with 95,475 mentions and 63,881 creators.

ALT Table comparing two requests: /topic/openai/v1 returning OpenAI Dev Day with 17 mentions and 14 creators, and /topic/open-ai/v1 returning OpenAI with 95,475 mentions and 63,881 creators.

1
1
217
Pull the topic list once and map your watchlist to real slugs before anything queries them. lunarcrush.com/en/developers…

Build with LunarCrush: API, MCP, and SDKs

Social sentiment, market data, creators, and trending topics for crypto and stocks — one API. MCP-ready for Claude, ChatGPT, and Gemini.

lunarcrush.com
19
posts_active and posts_created are different numbers. Only one means posts made that day. Friday, /topic/whatsapp/time-series/v1: 111,952 posts_active, 50,965 posts_created. Active counts older posts still collecting engagement. Label that axis wrong and you overstate by 2.2x.
Table of the whatsapp topic from Oct 19 to Oct 23 comparing posts_active and posts_created, with ratios from 1.7x to 2.2x, ending at 111,952 active against 50,965 created.

ALT Table of the whatsapp topic from Oct 19 to Oct 23 comparing posts_active and posts_created, with ratios from 1.7x to 2.2x, ending at 111,952 active against 50,965 created.

143
Research teams do not want a dashboard. They want a CSV. /topic/iphone/time-series/v1, bucket day, 30 days, one request. iPhone daily engagements sat near 400M for three weeks, then ran from 395.6M to 939.8M between Oct 20 and Oct 24. That shape only exists in history.
Bar chart of daily engagements on the iphone topic from Sept 27 to Oct 25, flat near 400M then rising to 940M on Oct 24, with a line marking that no day before Oct 22 cleared 520M.

ALT Bar chart of daily engagements on the iphone topic from Sept 27 to Oct 25, flat near 400M then rising to 940M on Oct 24, with a line marking that no day before Oct 22 cleared 520M.

1
1
178
Twelve lines. Set LUNARCRUSH_API_KEY, pass a topic, get a file your notebook can read. The newest row is a partial day, so it goes.
Twelve lines of Python that call the topic time-series endpoint with daily buckets, drop the partial day, and write a CSV of time, interactions and posts_created.

ALT Twelve lines of Python that call the topic time-series endpoint with daily buckets, drop the partial day, and write a CSV of time, interactions and posts_created.

11
Load more