For-you feed w/ external providers

To choose a different provider, simply pass in a different value in the provider field. openrank is set as the default. (karma3 is an older name for openrank --kept here for backwards compatiblity--)

If you pick mbd as provider, you can further customize your feed by passing in additional filter values in an optionalfilters object inside the provider_metadata field in the request e.g.

const provider_metadata = encodeURIComponent(JSON.stringify({
  "filters": {
    "channels": [
      "https://warpcast.com/~/channel/neynar"
    ],
    "languages": [
      "en"
    ],
    "author_ids": [
      "194",
      "191"
    ],
    // remove_author_fids only works when author_ids isn't passed in
    // "remove_author_ids": [
      // "18949"
    // ],
    "frames_only": false,
    "embed_domains": [
      "neynar.com",
      "frames.neynar.com"
    ],
    "ai_labels": [
      "science_technology"
    ],
		"remove_ai_labels": [
    	"spam"
    ]
  }
}));

The filters available for MBD that you can pass in that object are:

Name	Type	Description
`start_timestamp`	string	return only casts after this start_timestamp, specified as Epoch time (Unix timestamp)
`end_timestamp`	string	return only casts before this end_timestamp, specified as Epoch time (Unix timestamp)
`channels`	string[]	return only casts that belong to these channels, specified by channel urls (root_parent_url)
`languages`	string[]	returns only casts that use these languages
`author_ids`	string[]	returns only casts created by authors with these fids
`remove_author_ids`	string[]	does not return casts created by authors with these fid's NOTE: this is ignored if author_ids is defined
`frames_only`	boolean	whether to limit search to only frames
`embed_domains`	string[]	return only casts with specific domains embedded
`ai_labels`	string[]	Return only casts that have these AI labels. Available values below. Labels in topics category: - `arts_culture` - `business_entrepreneurs` - `celebrity_pop_culture` - `diaries_daily_life` - `family` - `fashion_style` - `film_tv_video` - `fitness_health` - `food_dining` - `gaming` - `learning_educational` - `music` - `news_social_concern` - `other_hobbies` - `relationships` - `science_technology` - `sports` - `travel_adventure` - `youth_student_life` Labels in sentiment category: - `positive` - `neutral` - `negative` Labels in emotion category: - `anger` - `anticipation` - `disgust` - `fear` - `joy` - `love` - `optimism` - `pessimism` - `sadness` - `surprise` - `trust` Labels in moderation category: - `llm_generated` - `spam` - `sexual` - `hate` - `violence` - `harassment` - `self_harm` - `sexual_minors` - `hate_threatening` - `violencegraphic` Labels in web3_topics category: - `web3_nft` - `web3_defi` - `web3_infra` - `web3_industry` - `web3_consumer`
`remove_ai_labels`	string[]	do not return casts with these AI labels NOTE: this is ignored if ai_labels is defined

A full request to the feed API with the custom mbd filters object looks like below

const fetch = require('node-fetch');


const provider_metadata = encodeURIComponent(JSON.stringify({
  "filters": {
    "channels": [
      "https://warpcast.com/~/channel/neynar"
    ],
    "languages": [
      "en"
    ],
    "author_ids": [
      "194",
      "191"
    ],
    // Note: remove_author_ids only works when author_ids isn't passed in
    // "remove_author_ids": [
    //   "18949"
    // ],
    "frames_only": false,
    "embed_domains": [
      "neynar.com",
      "frames.neynar.com"
    ],
    "ai_labels": [
      "science_technology"
    ]
  }
}));

const url = `https://api.neynar.com/v2/farcaster/feed/for_you?fid=3&viewer_fid=2&provider=mbd&limit=10&provider_metadata=${provider_metadata}`;

const options = {
  method: 'GET',
  headers: {
    'accept': 'application/json',
    'api_key': 'NEYNAR_API_DOCS'
  }
};

// Fetch request with the metadata and options
fetch(url, options)
  .then(response => response.json())
  .then(data => console.log(data))
  .catch(error => console.error('error:', error));