Tweets and Search Results for Your Agent, Without the X API Tier
Give your agent tweets, user timelines, and keyword search as pay-per-call TikHub endpoints on Monid, with no monthly X API tier.

Copy this line to your agent to pull tweets and keyword search results without an X API tier.
set up https://monid.ai/SKILL.md and use tikhub/twitter to fetch a keyword search timeline and one user profile
Your agent can read tweets, user timelines, and keyword search results as ordinary pay-per-call endpoints, billed only when a call runs and with no monthly X tier to sign up for. On Monid the X data lives behind TikHub endpoints that you discover for free, inspect for free, and pay for one call at a time from a single wallet you already use for the rest of your data providers.
X priced most teams out of its own API. The tiers moved from cheap to steep, the free lane shrank to almost nothing, and a project that just wants to watch a keyword or read one account's recent posts suddenly needs a seat-based plan sized for an enterprise. We think that is the wrong shape for an agent. An agent does not want a subscription. It wants to make a call, get JSON back, and move on.
TL;DR
- Read tweets, user profiles, and keyword search as pay-per-call TikHub endpoints on Monid, no X API tier and no per-vendor signup.
discoverandinspectare free. You only pay onrun, at the price shown before you commit.- The X data shares one wallet with every other provider on Monid, so social sits next to enrichment and web search in the same billing.
- Cost is per call, not per seat: a keyword search or a profile pull is a fraction of a cent to a few cents territory. See live pricing at monid.ai/tools.
- Monid ships as an MCP server, so the agent routes the call itself instead of you wiring an X client.
Why this is a product decision, not a workaround
Scraping X yourself means rotating sessions, tracking layout changes, and babysitting rate limits, all so your agent can answer one question. Buying an official tier means committing to a monthly floor before you know whether the data is even useful. Neither fits an agent that dips into social data occasionally and needs the read path to just work.
Monid is a pay-per-call data API marketplace: one interface and one wallet to discover and execute hundreds of external data endpoints (web scraping, people and company enrichment, social data, web search, generative media) without a separate signup per vendor. The X read surface is one of those endpoints. Below are the principles that shape how it behaves.
The principles behind it
1. Search and read are runtime calls, not a plan you buy
A keyword search should be something the agent does in the moment, the same way it calls any other tool. On Monid the TikHub search timeline endpoint (/api/v1/twitter/web/fetch_search_timeline) takes a query and returns matching tweets. The profile endpoint (/api/v1/twitter/web/fetch_user_profile) returns one account's public profile. Tweet-detail style endpoints round out the set for pulling a single post and its context. There is no plan to provision first. The first time your agent needs tweets, it discovers the endpoint and runs it.
2. One wallet across every social and data source
X data does not live in its own silo with its own invoice. It draws from the same balance as People Data Labs enrichment, Exa web search, Apify scrapers, and everything else in the catalog. When your agent enriches a company, searches the open web, and then checks what that company is tweeting, all three are one line item on one wallet. That is the difference between an agent that can freely combine sources and one that stalls at every vendor boundary.
For agents
Grab an API key at app.monid.ai, then paste this to your agent and hand it the key:
set up https://monid.ai/SKILL.md
It learns the whole discover, inspect, run workflow itself. More details in the agent quickstart.
For humans
npm install -g @monid-ai/cli
monid keys add --label main --key <your-api-key>
More details in the CLI quickstart.
3. Priced per call, not per seat
You pay for the calls you make, at the magnitude of a fraction of a cent to a few cents per result depending on the endpoint, and nothing when you are idle. A weekend project that pulls a hundred search results costs cents. There is no seat count, no monthly minimum, and no tier to outgrow. Discovery and inspection stay free, so you can read the exact schema and the exact price before a single paid call. Current numbers are always on monid.ai/tools.
Here is what a real run looks like. Discover and inspect first (both free), then run the search with a small limit:
monid discover -q "twitter keyword search tweets"
monid inspect -p tikhub -e /api/v1/twitter/web/fetch_search_timeline
monid run -p tikhub -e /api/v1/twitter/web/fetch_search_timeline -i '{"keyword":"monid","search_type":"Latest"}' --query '{"count":20}' -w
Swap in the profile endpoint the same way when you want one account instead of a keyword:
monid run -p tikhub -e /api/v1/twitter/web/fetch_user_profile -i '{"screen_name":"monid_ai"}' -w
4. MCP-native, so the agent routes it
Monid ships as an MCP server. The agent sees discover, inspect, and run as tools and decides for itself when tweets are the right data. You do not hand-wire an X client or teach the model an SDK. It finds the TikHub endpoint, reads the schema, and calls it, the same loop it uses for every other provider in the marketplace.

FAQ
Do I need an X or Twitter developer account? No. The X data comes through TikHub endpoints on Monid, so there is no X developer signup and no monthly X API tier. Your Monid key is the only credential.
What can the agent actually read?
Keyword and search timelines, a user's public profile, and single tweets through the tweet-detail style endpoints. Run monid discover -q "twitter" to see the current TikHub X endpoints and their exact request schemas.
How is this billed? Per call, pay-as-you-go, at the price shown when you inspect the endpoint. No seats, no minimums. Discovery and inspection are free. See magnitudes and live pricing at monid.ai/tools.
Can I combine tweets with other data in one workflow? Yes, that is the point of one wallet. The same agent can search tweets, enrich the people behind them, and run a web search, all from a single Monid balance and a single MCP connection.
Point your agent at monid.ai and give it the X read path it was missing, priced one call at a time.