mcpField
S
social-intelligence

social-intelligence

github.com/merit-systemsMarketing & growth23141 installsUpdated 10w ago
Repository

About social-intelligence

social-intelligence is a skill for accessing Reddit through x402-protected APIs. It searches posts and retrieves comments, supporting brand monitoring, competitor research, and community sentiment analysis.

Key features

  1. 01Search Reddit posts by query
  2. 02Sort and limit results by timeframe
  3. 03Scope searches to a subreddit
  4. 04Retrieve post comments with pagination
  5. 05Access post and comment metadata

Use cases

  1. 01Monitor brand mentions on Reddit
  2. 02Research competitor feedback and reviews
  3. 03Assess sentiment in community discussions

Install

Copy the command and run it in a terminal.

$npx skills add https://github.com/merit-systems/agentcash-skills --skill social-intelligence

After installing, type/social-intelligencein Claude Code to confirm it loaded.

Bundled filesrules/getting-started.mdrules/rate-limits.md

Details

From the project README

Social Intelligence with x402 APIs

Access Reddit through x402-protected endpoints.

Setup

See rules/getting-started.md for installation and wallet setup.

Quick Reference

TaskEndpointPriceDescription
Search Reddithttps://stableenrich.dev/api/reddit/search$0.02Search Reddit posts
Get commentshttps://stableenrich.dev/api/reddit/post-comments$0.02Comments on a post

See rules/rate-limits.md for usage guidance.

Reddit

Search Posts

Search Reddit for posts:

npx agentcash@latest fetch https://stableenrich.dev/api/reddit/search -m POST -b '{"query": "best programming languages 2024"}'

Parameters:

  • query - Search terms (required). Scope to a subreddit with Reddit's subreddit: operator inside the query (there is no separate subreddit field)
  • sort - relevance, new, top, comment_count (default: relevance)
  • timeframe - all, day, week, month, year (default: all)
  • maxResults - Maximum results, 1-25 (default: 10)
  • after - Pagination cursor from a previous response

Returns:

  • Post title and content
  • Author and subreddit
  • Upvotes and comment count
  • Post URL
Search in Subreddit
npx agentcash@latest fetch https://stableenrich.dev/api/reddit/search -m POST -b '{
  "query": "typescript vs javascript subreddit:programming",
  "sort": "top",
  "timeframe": "year"
}'
Get Post Comments

Get comments from a Reddit post:

npx agentcash@latest fetch https://stableenrich.dev/api/reddit/post-comments -m POST -b '{"url": "https://reddit.com/r/programming/comments/abc123/..."}'

Parameters:

  • url - Full Reddit post URL (required)
  • cursor - Pagination cursor for more comments

Returns:

  • Post details
  • Comment text and author
  • Comment scores
  • Comment timestamps
  • hasMore / cursor for pagination

Workflows

Standard
  • (Optional) Check balance: npx agentcash@latest balance
  • Use npx agentcash@latest discover https://stableenrich.dev to list endpoints
  • Use npx agentcash@latest check <endpoint-url> to see expected parameters and pricing
  • Call endpoint with npx agentcash@latest fetch
  • Parse and present results
Brand Monitoring
  • (Optional) Check balance: npx agentcash@latest balance
  • Search Reddit for discussions
  • Summarize sentiment and key mentions
npx agentcash@latest fetch https://stableenrich.dev/api/reddit/search -m POST -b '{"query": "YourBrand", "sort": "new"}'
Competitor Research
  • Search Reddit for competitor reviews
  • Analyze common complaints and praise
npx agentcash@latest fetch https://stableenrich.dev/api/reddit/search -m POST -b '{"query": "competitor name review", "sort": "top", "timeframe": "year"}'
Community Sentiment
  • Identify relevant subreddit
  • Search for discussions on topic
  • Get comments from top posts
  • Synthesize overall sentiment
npx agentcash@latest fetch https://stableenrich.dev/api/reddit/search -m POST -b '{"query": "new feature name subreddit:relevant_community", "sort": "top"}'

npx agentcash@latest fetch https://stableenrich.dev/api/reddit/post-comments -m POST -b '{"url": "https://reddit.com/..."}'

Response Data

Reddit Post Fields
  • title - Post title
  • selftext - Post body (for text posts; selftextTruncated flags truncation)
  • author - Username
  • subreddit - Subreddit name
  • score - Upvotes minus downvotes
  • numComments - Comment count
  • permalink - Link to post
  • createdAt - Timestamp

Search responses also include after (pagination cursor) and searchContext (query + result count).

Reddit Comment Fields
  • body - Comment text (bodyTruncated flags truncation)
  • author - Username
  • score - Net upvotes
  • createdAt - Timestamp

Comments are returned as a flat list; use hasMore and cursor to page through more.

Cost Estimation

TaskCallsCost
Reddit thread + comments2$0.04
Full monitoring scan3-5$0.06-0.10

Comments

Markdown supported