Collectors read each vendor's billing and the systems your work happens in. Your admin declares the facts no API holds: who is on which team, which login is which person. The engine joins the two, and every page states which sources it had.
Settled cost on one side. Finished work on the other.
Dashed systems are planned. Every integration and where it stands.
Read-only keys for the AI vendors you pay and the systems your teams work in. Amounts, people, dates and states. Never content.
The facts no API holds: who is on which team, which login is which person, what you committed to. Every declaration keeps its author.
Every page states which sources it had and how strong each link is, so a number is never stronger than its evidence.
Each install is assigned to a person, and each person's GitHub login is declared as an alias. Spend reaches the person through the first, merged pull requests through the second, and for each person and month the join sets settled spend beside merged work.
Both numbers stay on the page. Nothing blends them into a score, and a month missing either side is left out rather than guessed.
Collectors run on the schedule your admin sets, and each run is logged with its outcome. Here is what each source adds to the record.
Daily billed cost from Anthropic's cost report, in the currency it was billed in. It does not yet ask for the breakdown by model, so these rows read no model declared. Where the vendor sends no token reading, tokens read unavailable, not zero.
Daily billed cost and currency for the organization from OpenAI's Costs API, not yet broken down by project or key. It names no model, so OpenAI rows read no model declared instead of borrowing one.
Team spend from Cursor's Admin API, broken down per user, so Cursor cost lands on a person's page beside their other tools.
Copilot usage and billed cost from GitHub's billing usage report. It is the one collector that brings a vendor list price.
Pull requests and the commits in them: who merged what, when, and how many lines each pull request changed. Metadata only, never the contents of a file. Logins reach people through the aliases you declare.
Issues and epics, so cost can sit beside resolved work and be split across the projects people closed issues in. The split is stated as a share of cost, never as what a project used.
Groups and their members. Finding a group maps nobody; it joins a team only when an admin maps it. Okta sign-in for Business accounts is built and not yet switched on in our hosted service.
Your own billing exports, for spend no API reaches, to be filed in the same ledger beside provider billing and corrections. Upload is planned; each row will keep its source, so a figure built from it can name where it came from.
On each Mac: Claude Code session files, Claude desktop sessions, Codex's local database and session files, and Cursor plan usage.
Every source lands in one cost record shaped after the FOCUS billing specification, with the same columns whether ai-cur runs hosted, on your servers or on Databricks.
Some facts live in contracts, org charts and policy, not in any API. Your admin declares them on the admin page or from the command line, and every declaration records who made it and when.
Create teams and map each identity provider group to one. People who reach no mapped team are listed, so nobody drops out quietly.
Tie a GitHub login to a person, so the pull requests they merge land beside the spend on their seats and keys. Without an alias, merged work has nobody to sit beside.
Name who holds management, admin or finance. The database checks the role before it returns a single row, so a role nobody declared opens nothing.
A monthly amount for a team or the whole account, its alert thresholds, and soft and hard caps. A blank cap means none declared, not a cap at the full amount.
Budgets that cut across teams, for a project or a person, with caps of their own. A project budget counts each person in full, so these never add up to the account total.
The amount, currency and term of a contract with a provider. Drawdown counts settled charges inside that term only, and a commitment is never inferred from spend.
List prices per provider and model. Plan fit needs one before it can say anything, and no rate card ships with the product.
The smallest roster a team needs before it is compared with teams it is not run by. The roster counts everyone mapped to the team, active or not.
Which plan an account holds, on what term and until when. An admin declares it, and it lands in the signed entitlements the desktop client checks. Paying for a Team plan will declare the same thing once payments open.
Enroll an install to issue its token, shown once. Rotate it with a short overlap, revoke it at once, and assign each install to its person.
The desktop client records the repository, branch and task a session ran in, so the cost of that help attaches to the item itself.
One person's settled spend beside what that person finished in the same period, through the aliases you declare.
Where an individual link is not available or not appropriate, a team's spend sits beside what the team delivered.
Which lines an assistant wrote is a signal vendors and attribution tools already work on, and ai-cur does not rebuild it.
What it holds is simpler and checkable: one person's settled spend and the same person's merged pull requests, for the same month, side by side. It never credits a tool with the work, and it publishes no attribution accuracy figure because it makes no attribution claim.
Routing to cheaper models, caching and hard limits per key already exist in gateways you can run yourself. ai-cur stays out of the request path. It reads billing through each vendor's reporting API and never handles a model request, a prompt or a response.
It is built to read the whole bill, including seats no proxy ever sees, and serves the cap decision for each team, project or person at an endpoint your gateway or key issuer checks.
Declare your teams and vendors once, and every view above reads from that declaration instead of a guess.
