Serre Connect tool reference

Every tool your assistant can use once it is connected to Serre, with the exact arguments each one takes. This page is generated from the connection itself, so it always matches what your assistant actually sees.

Connect is in Beta. To set one up, see Connect an AI assistant.

Reads are available to every key. Writes need a full-access key. Every call runs as the member who owns the key, inside their workspace only.

Tool	Kind	Uses your allowance
list_value_drivers	read	no
list_solutions	read	no
list_prospects	read	no
get_prospect	read	no
create_prospect	write	no
create_business_case	write	no
generate_business_case	write	yes — the same as generating in the app
get_case_results	read	no
share_case	write	no — needs a plan with shareable links
plan_business_case	read	no
execute_case_plan	write	yes — it generates the case

list_value_drivers

Read. List the workspace's reusable value drivers. Use these ids in create_business_case so the generated case carries real figures.

No arguments.

list_solutions

Read. List this workspace's solutions with their linked value drivers. Use the ids when creating a business case.

No arguments.

list_prospects

Read. List the workspace's prospect companies (most recently updated first).

No arguments.

get_prospect

Read. Read one prospect's full buyer context.

Argument	Type	Required
prospectId	string	yes

create_prospect

Write. Create a prospect company. Duplicate detection by name or website is inherited from the workspace flow: a duplicate returns an error naming the existing prospect unless force is true.

Argument	Type	Required
companyName	string	yes
website	string	no
industry	string	no
dealContext	string	no
force	boolean	no

create_business_case

Write. Create a draft business case for a prospect. Name solution ids (from list_solutions) and/or value-driver ids to include. A case with no drivers generates to zero, so pass drivers when you want real figures.

Argument	Type	Required
prospectId	string	yes
title	string	no
selectedSolutionIds	string[]	no
selectedDriverIds	string[]	no

generate_business_case

Write. Generate the case: run the calculation engine and freeze the deliverable snapshot. Consumes 1 AI Generation and 1 credit from the workspace's ledger, the same metering as generating in the app.

Argument	Type	Required
caseId	string	yes

get_case_results

Read. Read a generated case's computed outputs (total benefit, ROI, payback, NPV, Monte-Carlo ranges) plus the workspace link and, if shared, the public URL.

Argument	Type	Required
caseId	string	yes

share_case

Write. Mint (or refresh) the case's public share link and return the URL a buyer can open without logging in. Uses the workspace's existing share machinery: expiry and regeneration inherited; revoking the key stops future minting.

Argument	Type	Required
caseId	string	yes

plan_business_case

Read. Plan a complete business case from a prospect WITHOUT changing anything. Returns status 'ready' with the plan, its cost, and a planToken for execute_case_plan; or 'needs_choice' (which prospect?) / 'needs_confirmation' (create this prospect?) with the question to put to the user. Ask the user, then call again with their answer (prospectId for a choice; createProspect: true to confirm creation). Give a website when you have one — it identifies a company more reliably than its name. In a multi-company workspace the case is sold by the default company unless companyId says otherwise.

Argument	Type	Required
companyId	string	no
prospect	string	no
website	string	no
prospectId	string	no
createProspect	boolean	no
solutionIds	string[]	no
driverIds	string[]	no
title	string	no
dealContext	string	no

execute_case_plan

Write. Run a plan from plan_business_case exactly as shown: creates the prospect if the plan said so, creates the draft case, generates it. Consumes 1 credit and 1 AI Generation. Only call this after the user has confirmed the plan.

Argument	Type	Required
planToken	string	yes

Errors

A call fails in one of two shapes, and the difference is deliberate:

• A protocol error (error on the JSON-RPC envelope) means the request itself was wrong: an unknown tool, a method the server does not have, a missing key.
• A tool failure (result.isError: true with a message in content) means the tool ran and could not do what was asked: invalid arguments, a duplicate prospect, a plan token that expired, too many requests. The message is written for the assistant to read back and act on.

Limits

Calls are capped per workspace per minute. Over the cap, a call comes back as a tool failure asking the client to wait a moment. Generating a case is additionally subject to the workspace's AI allowance, the same way it is in the app.