Server URL
The fastest way to connect is the in-app installer:
promptloop.com/mcp/install has one-click
buttons and copy-paste configuration for every client.
Connect your client
- Cursor
- Claude
- ChatGPT
- Codex
- Claude Code
Add this to Cursor shows a Needs login prompt — click it to authorize with OAuth.
~/.cursor/mcp.json:Available tools
Finding work emails
find_work_emails runs the same contact enrichment as in-app Email Lookup. Give it the whole
file in one call, using one of these inputs:
Files can have up to 10,000 rows. Larger files are split internally into batches of at most 2,000
people (10,000 credits), but you get back a single
job_id with one progress view and one results
file. Headers are detected automatically, including common variants such as First Name,
Surname, Name, Job Title, Agency Name, Website, Domain, URL, LinkedIn, Email, and
CRM ID. Columns with unhelpful headers are recognized from their values. The response reports the
column_mapping used and any unmapped_columns. Pass columns to override it, for example
{"full_name": "Contact", "website": "Agency URL"}.
Each person needs a name and a way to identify their company:
Run with
dry_run: true first. It returns row counts, rejected rows with reasons, duplicates, the
column mapping, the maximum credit cost, your available credits, and can_launch, without charging
anything. People repeated in a file are looked up once and charged once.
Poll get_work_email_results with the job_id for progress and per-person results in the original
row order, with email, verified, confidence, email_status, and source. verified is
true only when the provider confirmed the mailbox is deliverable; confidence: "probable" is a
likely address that could not be confirmed. Pass format: "csv" for a download link (valid for 24
hours) to a CSV with your original columns plus email, verified, confidence, email_status,
source, status, and note. If one of those column names already exists in your file, the
result column is prefixed with promptloop_.
Each email found costs 5 credits, the same as in the app. People with no email found, rejected and
duplicate rows, failed lookups, and repeat lookups served from your team’s 30-day cache are free.
Submitting the same file again within 24 hours returns the same job instead of starting a new one.
If rows fail, for example while the email data provider is temporarily unavailable,
retry_work_emails reruns only those rows. Free plans can’t use these tools.
HubSpot utilities for admins
The following tools are advertised only when the person who authorized the MCP OAuth connection is currently a PromptLoop admin with user mode disabled in the connected team, and that team has HubSpot connected. This is the PromptLoop admin role, not a HubSpot portal role. Ordinary team API keys do not grant this access. Authorization is checked again on every tool call, and HubSpot’s own connection permissions still apply.
Dataset tools take
dataset_id, file_id, and domain_column. file_id is the saved
CSV’s fileId, not the dataset version’s datasetFileId. Both the dataset and file must
belong to the connected team, and the file must be linked to that dataset.
Inspect properties and preview matching before syncing. The preview shows domain matching
and row decisions, not exact property changes. Sync defaults to update-only and filling
empty properties, skips ambiguous domain matches, and never creates HubSpot properties.
Set overwrite_existing: true only when replacing existing values is part of the user’s
authorized scope. A timed-out sync may still be running; do not automatically retry it.
Setup prompt
Paste this into your assistant so it knows how to drive PromptLoop:Prompt template
