Solution · For developers

# You run agents all day, and each new session forgets. *Pay for less, and keep your project facts.*

Context Mode runs between your agent and the model. It sends the model the failing test, not the ones that passed, and keeps your early details all day. Your npm test fix loop cost 22% less in paired runs.

[Get started free](https://console.context-mode.com/login) [Read the docs](https://context-mode.com/docs/quick-start)

Your day

npm test Hour six cat .env

Your agent runs 600 tests. 1 fails.

Printed **17.0 KB** Sent to the model **1.2 KB**

[npm test fix loop · Claude Opus 5.5 **$0.087 → $0.068** *22% lower* 15 of 16 pairs cost less](https://context-mode.com/docs/benchmarks)

Late in a long session, you ask about the morning.

Direct **4 of 10** Context Mode **9 of 10**

Exact answers about early details, 200K window. Direct compacted 3 times; through Context Mode, 0.

Your agent tries to read your keys.

Blocked · never reaches your shell **Read environment files**

.env and .env.*. Examples and templates are allowed.

Protect is off until you turn it on. You pick the rules.

The problem

## You pay for every byte. *Then you explain it all again.*

**Long output, paid again**

A test run prints 17 KB. Your agent pays for every byte a tool prints, again on each later turn.

**Long sessions compact**

Early details get summarized away. The agent asks again, or guesses.

**New sessions start from zero**

You explain your deploy target and your stack each time.

How it works

## One command. *Keep your agent and your model.*

1. 01 **Run one command** `npx @context-mode/cli` signs you in and connects your agents. It works with Claude Code and Codex today; OpenCode and Pi are coming soon. See the [quick start](https://context-mode.com/docs/quick-start). Your own Anthropic or OpenAI account pays the model. We add no fee on tokens.
2. 02 **Each request gets smaller** Long tool output folds; errors stay in full. Older work goes to the archive, word for word.
3. 03 **The model reads less and remembers** Memory gives every agent the current facts, on any machine. Old details come back by lookup.

Your agent sends everything as usual. Context Mode folds long output, archives older work, adds your current facts and, if you turn on Protect, checks each tool call. The model gets a smaller request.

Proof

## The same task, run twice. *Lower cost, same result.*

**22% lower** cost of a short npm test fix loop on Claude Opus 5.5, $0.087 to $0.068, 15 of 16 pairs

**155 of 168** shorter paired runs cost less; task checks passed equally in both arms

**9 of 10** early details answered exactly in a long session on a 200K window with Recall, against 4 of 10 direct

**0 compactions** in that 200K window run, against 3 direct

| Task | Context Mode | Direct | Lower |
| --- | --- | --- | --- |
| Code search, grep over 150 files | $0.066 | $0.093 | 29% |
| npm test fix loop, short | $0.068 | $0.087 | 22% |
| One large log | $0.071 | $0.091 | 22% |

Paired runs, Claude Opus 5.5, cost per task at API list price. The memory and compaction figures come from the 200K window run on [Recall](https://context-mode.com/recall). For a long 30-step session, see [Cut agent spend](https://context-mode.com/solutions/cut-agent-spend). [Method and sources →](https://context-mode.com/docs/benchmarks)

Plans

## Start free. *Move up when it pays for itself.*

**Free**

1,000 requests, once. About a week of very heavy use. Every feature, no card.

**Starter**

10,000 requests a month. For developers who use agents most days.

**Pro**

50,000 requests a month. For developers who run agents all day.

Every feature on every plan. Nothing breaks at a limit: your agent keeps working. [See plans and prices →](https://context-mode.com/pricing)

Products used

## What this *uses.*

**[Context Saving](https://context-mode.com/context-saving)**

Pay for the failing test, not the ones that passed.

**[Recall](https://context-mode.com/recall)**

All-day sessions with no compaction. Older work is looked up word for word.

**[Memory](https://context-mode.com/memory)**

Facts from your sessions, at their current value, in every agent.

**[Protect](https://context-mode.com/protect)**

Keep .env files, main and your home folder safe. Off until you turn it on.

**[Quick start](https://context-mode.com/docs/quick-start)**

Connect your agents in one command.

**[Pricing](https://context-mode.com/pricing)**

Free, Starter and Pro. Every feature on each.

Related: [Long sessions](https://context-mode.com/solutions/long-sessions) · [For startups](https://context-mode.com/solutions/startups) · [Benchmarks](https://context-mode.com/docs/benchmarks)

## Keep your agent. *Pay for less of what it reads.*

*npx @context-mode/cli* connects your agents. 1,000 requests free, no card.

[Get started free →](https://console.context-mode.com/login) [Read the docs →](https://context-mode.com/docs/quick-start)