What did we spend on OpenAI last month, and which project costs most?
This connects your OpenAI organisation's Admin API for usage and cost. It never calls a model. Memtro reads daily costs by project and line item, usage by model and project, projects, project members and the names of project API keys, never any key value. When asked and once an administrator has granted the action by name, it can create or rename projects, invite people by email, change or remove members and roles, and change a project's rate limits, which affects live traffic at once. Archiving a project is permanent and stops its API keys working.
Ask in plain English from chat, your editor, a scheduled job or your own agent. The model picks the right OpenAI (usage and cost) tool, reads what it needs, and answers with the source cited.
These tools change something in OpenAI (usage and cost). The model is told to use them only when you ask for exactly that, and an administrator grants each one by name to the teams that may use it; a personal connection acts only after its owner allows it.
Turn events on in the connection's Events tab on the Connections page, then add the address it shows as a webhook in OpenAI (usage and cost), choosing the same kinds; the tab shows OpenAI (usage and cost)'s own steps. Each event can start a job, with the event as its input. OpenAI (usage and cost) signs every delivery, and one whose signature does not check out is refused before it is read.
Events and webhooksOnce OpenAI (usage and cost) is connected, these are the kinds of things people ask. Plain questions, no syntax. If the answer lives partly in another system you have connected, Memtro looks there too.
What did we spend on OpenAI last month?
Which project has cost most in the last 90 days?
How many tokens did each model use in the last week?
Which API keys has nobody used in a month?
Add Priya Sharma to the data-platform project as a member.
Sarah Chen, finance director at Acme Ltd, asked what the company spent on OpenAI last month and which project cost most. Memtro read the daily cost buckets totalled per project and named the most expensive one, in USD as OpenAI reports it.
Tom Reilly, platform engineer at Northwind, asked Memtro to remove a departing contractor from the OpenAI organisation. Memtro removed the member, who lost access to every project at once, and listed the keys they owned by name so Tom could review them.
The teams that lean on OpenAI (usage and cost) most, a question each of them asks, and what they connect beside it.
A record in OpenAI (usage and cost) is rarely the whole story. Connect the systems that hold the rest and Memtro checks them together, then remembers what it learned so the next question starts further along.
Using something else for the same job, or both? Each connector has its own tools; connect as many as you like.
UptimeRobot
Honeycomb
Cronitor
FireHydrant
StatusCake
GlitchTip
GLPI
Langfuse
LibreNMS
NetBox
Snipe-IT
Woodpecker CI
Zabbix
LangSmith
Helicone
GrowthBook
Drata
Intruder
GitGuardian
Secureframe
ilert
JumpCloud
ConfigCat
Split
Aikido Security
Astronomer
Comet Opik
CrowdStrike Falcon
Detectify
Domotz
LambdaTest
LangWatch
LlamaCloud
Logz.io
Lunary
NinjaOne
Northflank
Portkey
PromptLayer
PRTG Network Monitor
Raygun
Retell AI
Runpod
Shodan
Spike.sh
Squadcast
Tenable
UpGuard
Uptime.com
Vanta
Zenduty
Cortex
WorkOS
Kandji
Automox
Huntress
SimpleMDM
Healthchecks.io
Rapid7 InsightVM
ZenhubIn OpenAI, open platform.openai.com → Settings → Organization → Admin keys. An ordinary standard or project API key will not work here.
Create a new Admin key named for Memtro, choosing read-only permissions if the dashboard offers them. A read-only key keeps the connection read-only. Copy it, as it is shown once.
In Memtro, open Connections → OpenAI (usage and cost) and paste it into Admin key. Memtro never calls a model with it.
Ask: "What did we spend on OpenAI last month?"
Chat in the dashboard, the OpenAI-compatible API, scheduled jobs, and any MCP client: Claude Code, OpenCode, Cursor, Claude Desktop. The OpenAI (usage and cost) tools are the same everywhere.
Only when you ask, and only what it has been granted: create an OpenAI project; rename an OpenAI project; archive an OpenAI project permanently: its API keys stop working; invite someone to the OpenAI organisation by email with a chosen role (OpenAI emails them); cancel a pending OpenAI organisation invite; change a member's OpenAI organisation role (owner or reader); remove a member from the OpenAI organisation; add an organisation member to an OpenAI project with a chosen role; change a member's role in an OpenAI project; remove a member from an OpenAI project; change an OpenAI project's rate limits for one model. Every other tool reads, searches or lists. The Connections page labels the tools that act, an administrator grants each one by name to the teams that may use it, and a personal connection acts only after its owner allows it.
A personal connection serves only your own requests. A shared connection serves the organisation: everyone, or only the teams an admin chooses. Admins can always add shared connections; other members need the permission.
Credentials are encrypted at rest. Tool results are cached briefly so agent loops do not repeat identical calls, and the facts the model extracts are saved to your organisation's memory with provenance (system, record and link) so they can be checked and superseded.
Admin key. In OpenAI, open platform.openai.com → Settings → Organization → Admin keys. An ordinary standard or project API key will not work here.
Yes. Turn events on in the connection's Events tab on the Connections page, then add the address it shows as a webhook in OpenAI (usage and cost), choosing the same kinds; the tab shows OpenAI (usage and cost)'s own steps. A job set to start "When an event arrives" then runs once for each matching event, with the event as its input. The run treats the event as outside data, never as instructions, and anything the run remembers is kept as unverified.
Seven-day free trial, then one monthly price per organisation. Connectors, jobs and memory are never metered.
Sign in, add a provider key, create an API key and point your tools at Memtro. It takes about five minutes.
7-day free trial · card on file, nothing charged until it ends · cancel any time
