New release v2026.626.0: Hermes agents built in, task watchdogs, and ask work mode.
See what's newAI spend should look like payroll, not like a surprise. Paperclip gives each agent a budget with a hard cap, meters every task, and rolls costs up in real time — overruns are impossible, not just unlikely.
A budget in Paperclip is enforcement, not a dashboard. When an agent reaches its cap it stops, marks its work with a clear state, and asks for more — it cannot quietly keep spending. Set caps per agent, per team, or per project.
Publish “Comparing agent runtimes” to the blog
Costs attach to tasks, tasks attach to goals — so you can answer "what did the migration cost us in AI spend?" the same way you’d answer it for cloud spend. By agent, by model, by week, by outcome.
Market brief: what changed since Monday
— Competitor X shipped usage-based pricing [source]
— Two enterprise threads mention agent budgets [source]
— Regulatory draft affects EU deployments [source]
3 claims verified · 1 refuted and omitted
Because roles and adapters are separate, you can staff mechanical work on inexpensive models and reserve frontier models for judgment. The budget page tells you when a role is over- or under-provisioned for what it actually does.
“AI spend should look like payroll: a number you set, a cap that actually holds, and a receipt attached to every piece of work.”
Budgets, caps, and receipts from the first agent you hire.