Integration · Company data

Connect Claude & ChatGPT to Faraday

Ask your AI which cohorts you have, how well a prediction performs, or to look up the predictions for one customer in Faraday. It can create and update cohorts and outcomes, build persona sets, set up pipelines and targets, and read datasets, connections, recommenders and traits. You can check and adjust predictions without clicking through every screen.

Connect Faraday free

Free plan, no credit card. Takes about a minute.

Try asking

“How good is our churn prediction?”
The AI reads the outcome's analysis, with model quality, lift and top predictors.
“Create a cohort of customers who ordered at least twice in the last year.”
It creates a cohort from your order event stream with a minimum count and a recency condition.
“Build a prediction of which leads will become customers.”
It creates an outcome with your customers cohort as the goal and your leads cohort as the eligible group.

What your AI can do

Cohorts and data

List datasets, connections and traits, and create and update cohorts from your event streams.

Predictions

Create and update outcomes, read their analysis, build persona sets and read recommenders.

Pipelines and delivery

Create pipelines and targets to deliver results, and look up one person's predictions through a lookup target.

“Split our customers into personas.”
It creates a persona set from the customer cohort, and Faraday groups the cohort into personas.
“What are the predictions for [email protected]?”
It looks up the person through a lookup target and returns their predictions.
“How has the size of the high-value cohort changed?”
It reads the cohort's membership counts over time.

28 tools for Faraday

These tools are switched on when you connect. You can switch any of them off, or require your approval before it runs.

create_cohortChanges dataCreate a cohort. Requires name; define membership with stream_name plus optional stream_conditions, recency, min_count/max_count, min_value/max_value, traits and place_conditions.
create_outcomeChanges dataCreate an outcome (a prediction). Requires name and attainment_cohort_id (who reached the goal); optional eligible_cohort_id, attrition_cohort_id, prediction_mode and bias_mitigation.
create_persona_setChanges dataCreate a persona set that groups a cohort into personas. Requires name and cohort_id; optional number_of_clusters and modeling fields.
create_pipelineChanges dataCreate a pipeline (scope). Requires name, population.cohort_ids (who to score), payload (which outcome_ids, persona_set_ids, cohort_ids, recommender_ids to include) and preview (true for a free preview).
create_targetChanges dataCreate a target to deliver a pipeline's results. Requires name, scope_id, representation and options; optional connection_id, filter and limit.
get_account_usageShow usage metrics for your current Faraday account.
get_cohortGet one cohort's definition and status by cohort_id.
get_cohort_membershipGet the membership counts over time for a cohort by cohort_id.
get_current_accountGet your Faraday account's details. Use to confirm which account you are working in.
get_datasetGet one dataset's details and status by dataset_id.
get_outcomeGet one outcome's definition and status by outcome_id.
get_outcome_analysisGet an outcome's performance analysis (model quality, lift, top predictors) by outcome_id.
get_persona_setGet one persona set and its personas by persona_set_id.
get_pipelineGet one pipeline (scope) by scope_id.
get_recommenderGet one recommender's details by recommender_id.
get_targetGet one target's details by target_id.
list_cohortsList your cohorts, the groups of people defined from your data (e.g. customers, leads, churned).
list_connectionsList your data connections (warehouses, buckets and other sources and destinations).
list_datasetsList your datasets, the customer data you have brought into Faraday.
list_outcomesList your outcomes, the predictions of who will reach a goal (e.g. convert, churn).
list_persona_setsList your persona sets, the customer segments Faraday found in a cohort.
list_pipelinesList your pipelines (scopes), which combine a population with the predictions to deliver.
list_recommendersList your recommenders, the next-best-product or offer predictions.
list_targetsList your targets, the deployment outputs of pipelines (exports and lookup APIs).
list_traitsList the traits (attributes about people) available in your account.
lookup_personChanges dataLook up one person's predictions through a lookup target by target_id, using email, phone, name and address, or latitude/longitude.
update_cohortChanges dataUpdate a cohort by cohort_id: name or any of its membership conditions. Only the fields you pass change.
update_outcomeChanges dataUpdate an outcome by outcome_id: name, cohorts, prediction_mode or bias settings. Only the fields you pass change.

Set up in three steps

1
Pick the app
Create a free PipMCP account and choose Faraday from the app list.
2
Paste your key
Paste your Faraday API key. You find it in the Faraday dashboard. PipMCP sends it as a Bearer token.
3
Add the link to your AI
You get a personal MCP link. Add it to Claude, ChatGPT or Cursor:
1. Click your name, then Settings › Connectors › Add custom connector.
2. Paste your link as the Remote MCP server URL.
3. Switch it on from the + menu in a chat.

Questions

What can the AI do in Faraday?

It can read your account, usage, connections, datasets, cohorts, outcomes, persona sets, recommenders, pipelines, targets and traits. It can create and update cohorts and outcomes, create persona sets, pipelines and targets, and look up a person's predictions through a lookup target. The tools do not include deleting or archiving anything.

Does the AI see my Faraday API key?

No. Your Faraday credentials are encrypted at rest and never shown to the AI. After you save them, they are not shown again, not even to you. The AI only sees the results of the tools it calls.

Can I control what the AI is allowed to do?

Yes. You choose which tools are switched on, so you can start read-only. Creating cohorts, outcomes, pipelines and targets changes your Faraday account, and each of these can require your approval before it runs. Every tool call is logged.

Does it work with ChatGPT?

Yes. In ChatGPT go to Settings › Apps & Connectors › Advanced and turn on Developer mode, then add your PipMCP link. Developer mode needs a paid ChatGPT plan: Plus, Pro, Business or Enterprise. The same link also works in Claude (Settings › Connectors › Add custom connector), Cursor and other MCP clients.

What does it cost?

PipMCP has a free plan with no credit card required. Paid plans bill per completed task. You also need your own Faraday account.

What is a pipeline in Faraday?

In the Faraday API a pipeline is called a scope. It combines a population, the cohorts you want to score, with the predictions to include: outcomes, persona sets, cohorts and recommenders. A target then delivers the results, for example as an export or a lookup API. When the AI creates a pipeline, it can mark it as a free preview.