Per-agent budget

Keys and accounts are deployment artifacts; agents are the units that decide to spend. When five agents share one key, a key-level cap punishes all five for the behaviour of one, and a key-level dashboard cannot say which one. Scoping the envelope to the agent makes both the limit and the accountability match the thing actually making decisions.

A summariser agent gets $20 a month and a research agent $200, on the same key. The research agent exhausting its envelope stops (or is throttled, or alerts — whichever action its budget declares) while the summariser continues untouched.

Utilisation is attributed spend over the envelope for the period. The mechanics that make it possible are attribution — every call carrying an agent identity — plus period resets; Capsera scopes budgets to an agent, a team, or the organisation over daily, weekly, or monthly windows.

What is a per-agent budget?

A per-agent budget is a spending envelope scoped to one named agent, rather than to the API key, project, or account the agent shares with everything else. Keys are deployment artifacts; agents are the units that decide to spend. Scoping the envelope to the agent makes both the limit and the accountability match the thing actually making decisions.

How do I set an AI budget for a team?

Create a budget with team scope and give it a period — daily, weekly, or monthly with calendar-aware resets — then choose what happens at the limit: alert, throttle, or block. Team budgets compose with per-agent ones, so a single agent can be capped inside a wider team allowance. Utilisation is attributed spend over the envelope for the period, which requires every call to carry a team identifier.

How do I set a budget for a team of agents?

Use a team-scoped budget so every agent carrying that team identifier draws from one envelope, and add per-agent budgets underneath for any agent that should not be able to consume the whole allowance. The team budget bounds the group's blast radius; the per-agent budgets bound each member's. Both are checked before the provider call, so neither reports an overrun after the fact.

• Agentic spend governance — Agentic spend governance is the practice of attributing, budgeting, and enforcing LLM spend at the level of the agents that generate it, rather than at the level of API keys or accounts.
• Pre-call enforcement — Pre-call enforcement is a budget check that runs before an LLM request is sent to the provider, so an over-budget call is stopped before it costs anything rather than reported after it has.
• Cost attribution — Cost attribution is the assignment of every unit of LLM spend to the agent, team, feature, or customer that caused it, rather than leaving it aggregated under a shared API key.