Solution
You have a few engineers on agents and a receipt for each. Start free, then put the team on one bill.
Each account starts with 1,000 free requests and every feature, no card. When the team is ready, Team puts every seat on one pool of requests and one bill.
We want to try it before we pay.
npx @context-mode/cli
signs you in and connects your agents
Free: 1,000 requests, once
every feature, no cardAbout a week of very heavy use. Your agents keep working when it runs out.
Five engineers, five plans, five receipts.
Pro, per person
- 50,000 requests a month
- Monthly or yearly, one account each
Team
- Per seat, talk to us
- 50,000 a month per seat, one pool
- One bill to the team owner, 2 seats or more
Your own Anthropic or OpenAI account pays the model. We add no fee on tokens.
The team used its pool. What breaks?
Your agent keeps working, with no error and no refused request.
Requests go straight to the model until next month.The team owner sets the budget. Protect's Core protections stay on, and your memory and rules are kept.
A new hire's agent runs cat .env.
Set once by the org. An agent cannot loosen it.
Blocked by your team’s policy: Read environment files. .env files hold secrets.
On Team, one Protect policy covers every member.
Each engineer sets up agents alone. Each with a bill and a rule set.
Each engineer buys a plan, so the founder adds up receipts to see what agents cost.
Agent permissions live in each client, on each machine. A new hire sets them up again.
Long test logs and big files go to the model in full, and you pay for every token.
One command per engineer. One place for the rules.
- 01Start free
Each engineer tries it on a free account: 1,000 requests, every feature, no card. Nothing to host.
- 02Move to Team
Write to sales@context-mode.com. Team starts at 2 seats.
- 03One bill, one pool
Every seat draws from one pool, 50,000 requests a month for each seat. The team owner gets one bill and sets the budget.
Try it on real work. Then decide.
Paired runs, the same task run both ways: Claude Code direct and through Context Mode, each with a plan written before the first pair. For a long session, see Cut agent spend. Method and sources →
What changes when you move to Team. Same features.
| Team | Pro, one per engineer | |
|---|---|---|
| Price | Per seat, talk to us | A Pro plan for each person, see Pricing |
| Requests | 50,000 a month per seat, one pool | 50,000 a month for each person |
| Bill | One bill to the team owner, 2 seats or more | Monthly or yearly, one account each |
| Protect | One policy for the org; an agent cannot loosen it | Each person's own rules |
| Memory | Each member's memory stays their own | Yours |
| Budget | Set by the team owner | You set it |
| At the limit | Paused for the team until next month | Paused until next month, or a budget you set |
From Pricing and Plans and limits. At a limit your agent keeps working: requests go straight to the model, and Protect's Core protections stay on.
The products a small team uses.
Folds test, build and log output before the model reads it. Errors stay in full.
Checks every tool call before your terminal runs it. One policy for the whole org.
Facts from your sessions reach every agent, on any machine. Each member's memory stays their own.
From 2 seats: one bill to the team owner, every feature on every seat.
Set it up once for the whole team. Start free today.
npx @context-mode/cli connects your agents. 1,000 requests free, no card.