To let an assistant use these tools, grant Insurance operations on the consent screen, or just
Sales pipeline under its Fine-tune. The scope is pipeline.
Read tools work with Read only or Read & write access; write tools only with
Read & write.
Your assistant doesn't see these tools straight away. It finds them with search_tools or
describe_tools, and runs one with use_tool, passing its name and
arguments. Most assistants do this for you: just ask for what you want.
What your assistant is told about this group
Read the sales pipeline: policy applications (status, stage, risk factors) and the quotes issued (premium, underwriting decision, and whether the customer can buy it), plus quick counts across the whole book. Customers buy through the checkout. get_quote_calculation shows how a quote's premium was worked out, and list_insurance_analytics_records lists the records behind an analytics figure. request_medical_evidence records that underwriting needs a GP report or a test, and record_gp_report_consent the customer's consent to a GP report the team took by phone.
The descriptions below are the ones your assistant reads, so they're written for it. When a connection covers more than one
organisation, the assistant passes org (the organisation's slug, from list_orgs) to say which one.
get_insurance_overview
Read Get insurance overview
Fast counts across the org's book: products, pipeline, policies (total + live), customers, open claims and treaties.
Arguments
| Field | Type | Description |
|---|---|---|
| org | string | Organisation slug: required when this authorization covers more than one organisation (see list_orgs) |
get_insurance_analytics
Read Insurance book analytics
The full book-of-business dashboard in one call: headline stock/flow numbers with a prior-window comparison, the book/underwriting/persistency/reinsurance metric families, distributions (product, cover type, age, sum assured, smoker, decisions, cancellation reasons), daily new-business timeline, persistency cohorts and the in-force trend. Money is in pence; ratios are 0-1 (null when undefined). period: today, week, month (so far), last7, last30, last90, last_month (last30 by default), or custom from/to.
Arguments
| Field | Type | Description |
|---|---|---|
| period | string |
One of: "today", "week", "month", "last7", "last30", "last90", "last_month", "custom"
|
| from | string | YYYY-MM-DD (implies custom) |
| to | string | YYYY-MM-DD (implies custom) |
| product_id | string | Narrow to one product |
| org | string | Organisation slug: required when this authorization covers more than one organisation (see list_orgs) |
list_insurance_analytics_records
Read Insurance analytics drill-down
The records behind one get_insurance_analytics metric, as the dashboard drill page lists them: policies (in_force, new_business, cancellations, cooling_off, rated), premium payments (premium_collected, premium_failed) or reinsurance cessions (cessions_unemitted, cessions_unacked). Same period and product scoping as the analytics tool; compliance-check fixtures excluded. total counts every matching record; at most 200 are returned.
Arguments
| Field | Type | Description |
|---|---|---|
| metric required | string |
One of: "in_force", "new_business", "cancellations", "cooling_off", "rated", "premium_collected", "premium_failed", "cessions_unemitted", "cessions_unacked"
|
| period | string |
One of: "today", "week", "month", "last7", "last30", "last90", "last_month", "custom"
|
| from | string | YYYY-MM-DD (implies custom) |
| to | string | YYYY-MM-DD (implies custom) |
| product_id | string | Narrow to one product |
| limit | integer | |
| org | string | Organisation slug: required when this authorization covers more than one organisation (see list_orgs) |
list_applications
Read List applications
The sales pipeline: policy applications, newest first. status: "referred" is the manual-underwriting referral queue: follow up with get_application for the referral reasons. Each row carries a milestones summary (achieved, required and missed counts, and the terminal milestone that ended the sale); milestones: "missing_required" lists sales that ended with a required milestone missed (the audits list's flag), "complete" those that ended with every required milestone met. Follow up with get_sale_milestones for the milestone-by-milestone verdict.
Arguments
| Field | Type | Description |
|---|---|---|
| status | string |
One of: "started", "in_capture", "quoted", "referred", "bound", "declined", "postponed", "abandoned", "anonymous", "handed_off"
|
| product_id | string | |
| milestones | string |
missing_required: a required milestone was missed; complete: none were
One of: "missing_required", "complete"
|
| limit | integer | Default 25, cap 100 |
| org | string | Organisation slug: required when this authorization covers more than one organisation (see list_orgs) |
get_application
Read Get an application
One policy application in full: captured risk factors, pipeline state and the underwriting trace (rule executions with referral/decline reasons): the place to see WHY a case was referred or declined.
Arguments
| Field | Type | Description |
|---|---|---|
| id required | string | |
| org | string | Organisation slug: required when this authorization covers more than one organisation (see list_orgs) |
list_quotes
Read List quotes
Issued quotes, newest first, filterable by status, underwriting decision, product, application and reference.
Arguments
| Field | Type | Description |
|---|---|---|
| status | string |
One of: "issued", "superseded", "expired", "accepted", "rejected"
|
| decision | string |
One of: "accept_standard", "accept_rated", "mixed"
|
| product_id | string | |
| application_id | string | Only quotes issued on this application |
| reference | string | Substring match |
| limit | integer | Default 25, cap 100 |
| org | string | Organisation slug: required when this authorization covers more than one organisation (see list_orgs) |
get_quote
Read Get a quote
One quote with premium, decision and bindability.
Arguments
| Field | Type | Description |
|---|---|---|
| id required | string | |
| org | string | Organisation slug: required when this authorization covers more than one organisation (see list_orgs) |
get_quote_calculation
Read Get a quote's premium calculation
The premium breakdown behind one quote, as the dashboard quote page shows it: per priced component the effective rate-table inputs (assumed_keys were defaulted by the engine), base rate per mille and its table, decision-table loadings, rating, journey and office loadings, and the wholesale annual/monthly and loaded monthly figures (a fixed-price tier component carries fixed_tier instead: the tier's price is the price); then policy fee, multi-benefit discount and total. Money is in pence; decimals are strings.
Arguments
| Field | Type | Description |
|---|---|---|
| id required | string | Quote id |
| org | string | Organisation slug: required when this authorization covers more than one organisation (see list_orgs) |
request_medical_evidence
Write Request medical evidence
Record that underwriting needs medical evidence on a policy application before it can go on: a GP report, a nurse screening or a test, with a note. A GP report then waits on the customer's consent (Access to Medical Reports Act), which the agent asks for in the chat, or record_gp_report_consent records when the team took it by phone. Confirm with the operator first. Managers only.
Arguments
| Field | Type | Description |
|---|---|---|
| application_id required | string | The policy application |
| evidence_type required | string |
One of: "gp_report", "nurse_screening", "hiv_test", "ecg", "blood_profile", "bespoke"
|
| note | string | Why it's needed, for the team (at most 1000 characters) |
| org | string | Organisation slug: required when this authorization covers more than one organisation (see list_orgs) |
record_gp_report_consent
Write Record a GP report consent
Record the customer's consent to a GP report the team took by phone or in person (Access to Medical Reports Act): they were told what the report is for and that they can see it before it is sent, ask their GP to correct it, or say no, and they agreed. Pass what they said, word for word. Covers every GP report on the application waiting on consent. A chat's consent is the agent's to record. Managers only.
Arguments
| Field | Type | Description |
|---|---|---|
| application_id required | string | The policy application |
| customer_reply required | string | What the customer said when they agreed, word for word |
| org | string | Organisation slug: required when this authorization covers more than one organisation (see list_orgs) |