Purefeed API

Connect your AI tools directly to your Purefeed intelligence feed. Use the REST API or install the OpenClaw skill to access your signals, search tweets, generate content, and publish — all from your AI assistant.

Quick Start

Get your API key

Go to purefeed.ai/profile and create an API key in the Public API Keys section.

Test your key

Terminal

curl -s https://purefeed.ai/api/v1/auth/me \
  -H "Authorization: Bearer YOUR_API_KEY"

Start using the API

Use the REST endpoints below directly, or install the OpenClaw skill for the easiest AI integration.

OpenClaw Setup

The fastest way to connect Purefeed to your AI assistant. One command installs everything — no MCP server, no npm packages, no Node.js required.

Terminal

# Install the Purefeed skill
clawhub install purefeed

# Set your API key
openclaw config set skills.purefeed.env.PUREFEED_API_KEY "pf_live_YOUR_KEY_HERE"

That's it. OpenClaw will automatically use the Purefeed API when you ask about your Twitter feed, signals, or content generation. The skill teaches your AI assistant how to call the REST API directly via curl.

Don't have OpenClaw? You can use the REST API directly from any tool or script. All endpoints are documented below with curl examples.

API Reference

Base URL: https://purefeed.ai/api/v1

All requests require an Authorization: Bearer YOUR_API_KEY header. Responses return { "data": ..., "error": null } on success.

Account

GET/auth/me

Get the authenticated user's profile. Use to verify your API key works.

Feed & Discovery

GET/feed

Retrieve tweets ranked by signal relevance.

limitint

Max number of tweets to return

searchstring

Filter tweets by text search

signal_idsstring

Comma-separated signal IDs to filter by

POST/search

Full-text search across your signal-matched tweets.

querystringrequired

Natural language search query

limitint

Max number of results

GET/feed/signals

Get AI signal analysis for specific tweet IDs.

tweet_idsstringrequired

Comma-separated tweet IDs to analyze

Folders

GET/folders

List all bookmark folders.

GET/folders/:id/tweets

Get tweets saved in a specific folder.

limitint

Max number of tweets to return

Signals

GET/signals

List all signal configurations.

POST/signals

Create and auto-activate a new signal. Takes up to 6 hours to process existing tweets.

namestringrequired

Name for the signal

descriptionstring

Natural language description of what to detect

tagsstring[]

Tags for categorization

include_keywordsstring[]

Keywords to include in matching

exclude_keywordsstring[]

Keywords to exclude from matching

cronstring

Cron schedule (default: "0 */6 * * *"). Use "* * * * *" for realtime

timezonestring

IANA timezone for the schedule (default: "UTC")

GET/signals/:id

Get details of a specific signal.

PUT/signals/:id

Update a signal's name, description, or schedule.

namestring

New name

descriptionstring

New description

cronstring

New cron schedule (e.g. "0 5 * * 4" for Thursdays at 5am)

timezonestring

IANA timezone (e.g. "Europe/Madrid")

DELETE/signals/:id

Permanently delete a signal (irreversible).

GET/signals/:id/matches

Get tweets matching a specific signal.

limitint

Max number of matches to return

Content Studio

GET/styles

List available writing styles. Call before generating posts.

POST/studio/generate

Generate a tweet or thread using a writing style. Returns streaming text.

styleIdintrequired

Writing style ID (from /styles)

tweetIdsstring[]

Tweet IDs to use as inspiration

ideaPromptstring

Free-form prompt describing what to write about

POST/studio/humanize

Score a post for AI-likeness (1-100) and get a rewrite if score > 30. Returns streaming text.

postsstring[]required

Array of post texts to evaluate

POST/studio/publish

Publish a post to X/Twitter immediately. Requires Typefully connection.

postsstring[]required

Array of post texts (multiple = thread)

POST/studio/schedule

Schedule a post for a specific time.

postsstring[]required

Array of post texts

scheduledAtstring (ISO 8601)required

When to publish

POST/studio/draft

Save a post as a draft in Typefully.

postsstring[]required

Array of post texts

POST/studio/queue

Add a post to the next available posting slot.

postsstring[]required

Array of post texts

Response Examples

Feed / Search Response

GET /feed

{
  "data": [
    {
      "tweet_id": "1234567890",
      "full_text": "The latest AI breakthroughs are...",
      "creation_date": "2026-03-10T14:30:00Z",
      "views": 12500,
      "replies": 42,
      "retweets": 189,
      "screen_name": "techleader",
      "profile_image_url": "https://pbs.twimg.com/...",
      "signals": [
        {
          "signal_name": "AI Research",
          "content": "Discusses new transformer architecture...",
          "sentiment": "positive",
          "industry_category": "Technology"
        }
      ]
    }
  ]
}

Signals List Response

GET /signals

{
  "data": [
    {
      "id": 1,
      "user_signal_name": "AI Research",
      "signal_description": "Track breakthroughs in AI...",
      "creation_date": "2026-01-15T09:00:00Z"
    }
  ]
}

Humanize Response (Pass)

POST /studio/humanize — score below threshold

{
  "data": {
    "score": 22,
    "reasons": "Natural tone, varied sentence structure...",
    "rewrite": null
  }
}

Humanize Response (Fail)

When the score exceeds 30, the response includes a rewritten version. The first line is JSON with the score and reasons, followed by the rewritten text.

POST /studio/humanize — score above threshold

{"score": 78, "reasons": "Repetitive structure, overly formal..."}
Here's a more natural version of your post that maintains
the core message while sounding more authentic...

Error Reference

Error	What to do
401 Unauthorized	API key is invalid or expired. Create a new key at purefeed.ai/profile.
429 Too Many Requests	Rate limit exceeded. Wait and retry (60 req/min).
No Typefully connection	Connect your Typefully account in settings to publish posts.
No LLM API keys configured	Add your LLM API keys in Studio settings.
Writing style not found	Call `GET /styles` to get valid style IDs.
Signal not found	Call `GET /signals` to get valid signal IDs.

Rate Limits

60 requests per minute

Rate limits are per API key.
Streaming endpoints count as 1 request each.
When rate-limited, the response includes a Retry-After header indicating how long to wait.
HTTP status 429 is returned when the limit is exceeded.

Need help? Reach out at team@purefeed.ai or check your profile settings for API key management.