Claude Code Weekly Limits Just Jumped 50% Through July 13: What Pro, Max, and Team Users Should Do With the Extra Quota

Anthropic raised Claude Code weekly usage limits by 50%, effective immediately through July 13 at 6PM PDT (1AM GMT, July 14). The increase is live on every paid plan: Pro, Max, Team, and seat-based Enterprise. It applies everywhere Claude Code runs: CLI, IDE extensions, desktop app, and web. There is nothing to enable; the higher ceiling is already on your account.

Try Apidog today

This increase stacks on top of last week’s doubled 5-hour limits. In practice, paid Claude Code users now have a 2x boost to short-window quota and a 1.5x boost to weekly quota at the same time, with no price increase during the temporary window.

This guide breaks down what changed, how to use the extra capacity in real engineering work, and how to apply it to API design and testing workflows with Apidog.

What changed in Claude Code limits

Claude Code uses two main rate limits on paid plans:

• 5-hour limits: a rolling cap on tokens used in any 5-hour window. This is the limit you hit during long coding sessions, refactors, or debugging runs. Anthropic doubled this limit last week.
• Weekly limits: a cap on total token usage across the full week. This is the limit that usually affects heavy users near the end of a sprint. This limit just increased by 50%.

Anthropic does not publish fixed token numbers for each tier and may tune them based on capacity. The practical impact is:

• Pro users get more headroom before hitting weekly usage warnings.
• Max users can sustain longer multi-day coding sessions on complex codebases.
• Team and seat-based Enterprise users get the same 50% lift per seat, which compounds across the team.

You can check your current usage in:

/usage

You can also view usage in the Claude Code IDE status bar or the account settings page on the web.

Why Anthropic is doing this now

There are two likely reasons.

First, Anthropic has been provisioning more compute through Q1 and Q2 of 2026. When additional capacity is available, raising limits is a direct way to increase usage and collect better data on how developers use Claude Code at scale.

Second, competition is increasing. Codex recently shipped its own /goal autonomous agent loop, and OpenAI has also been adjusting limits on developer plans. The temporary 50% increase gives teams less reason to split workloads across providers.

The important date is July 13 at 6PM PDT. Anthropic has not said whether the higher weekly ceiling will become permanent, revert to baseline, or change to a different level.

What the extra quota unlocks

The value is not just “more usage.” The higher limits change which workflows are practical.

1. Longer /goal agent runs

The doubled 5-hour limit plus the 50% weekly increase makes longer autonomous runs more viable.

Before this change, a complex refactor using the /goal autonomous loop could burn through the 5-hour budget and force you to pause. With both limits raised, you can give Claude Code a measurable objective and let it work longer without interruption.

Example goal:

/goal Refactor the billing service to use the new invoice schema.
Success criteria:
- All existing tests pass
- New schema is used by invoice creation and update flows
- No public API response fields are removed
- Add migration tests for legacy invoice records

This is especially useful for long-running tasks such as:

• Large refactors
• Test suite repair
• Dependency upgrades
• API contract implementation
• Monorepo cleanup

2. Larger codebase context

Claude Code is most useful when it has enough context to understand your project structure.

With more headroom, you can load larger working sets and reduce manual scoping. For example, instead of limiting Claude Code to one service directory, you can include related packages:

/services/billing
/packages/shared-types
/packages/api-client
/tests/integration/billing
/openapi

This helps with tasks that cross boundaries, such as:

• Updating shared DTOs
• Fixing contract mismatches
• Refactoring duplicated validation logic
• Tracing bugs across backend, SDK, and tests

If your monorepo previously felt too large for Claude Code, this temporary window is a good time to test that assumption again.

3. Multi-agent workflows

Multi-agent tools like Ruflo, the multi-agent orchestrator on top of Claude Code, run several Claude instances against the same task and merge their output.

These workflows consume quota quickly because you are effectively running multiple sessions in parallel. With the new limits, they become more practical for daily engineering work instead of occasional experiments.

Example workflow:

Agent 1: Inspect API handlers and identify contract mismatches
Agent 2: Review OpenAPI schema and generated client types
Agent 3: Add or repair integration tests
Agent 4: Merge findings and propose implementation plan

Use this when the task benefits from independent analysis, such as architecture review, large migrations, or test coverage gaps.

4. Heavier MCP server usage

Claude Code can call external tools through MCP, the Model Context Protocol. MCP calls count against quota, so chaining tools used to become expensive quickly.

With the higher ceiling, you can build workflows that use multiple MCP servers in sequence:

Claude Code
 -> GitHub MCP
 -> Database MCP
 -> API testing MCP
 -> Browser automation MCP
 -> Issue tracker MCP

Practical examples:

• Pull an issue from GitHub
• Inspect related code
• Query a test database
• Run API contract tests
• Open a browser automation check
• Push a fix and summarize the PR

If MCP setup fails because of configuration issues, see the invalid custom3p enterprise config fix.

How to use the next eight weeks

You have until July 13 to take advantage of the higher limits. Use that window intentionally.

1. Build the agent workflow you postponed

If you had a /goal workflow or multi-agent setup that was barely viable under the old limits, build it now.

Good candidates:

• Automated dependency upgrade flow
• Contract-first API implementation flow
• Test repair loop
• Large refactor workflow
• Issue-to-PR workflow

Even if limits return to baseline later, you will have real data on whether the workflow is worth upgrading plans for.

2. Move manual side-channel work into Claude Code

Use Claude Code for tasks you may have been doing manually or in cheaper tools:

• Writing AGENTS.md files
• Reviewing pull requests
• Generating OpenAPI specs
• Configuring MCP servers
• Creating test cases
• Explaining legacy modules
• Updating internal docs

For one practical starting point, see this guide to writing AGENTS.md files.

3. Stress-test your real workflow

Do not ration usage during the temporary window. Run Claude Code as if the new limits were permanent.

Track:

• How often you hit /usage warnings
• Which tasks consume the most quota
• Whether Max would be worth it over Pro
• Whether your team needs more seats
• Which workflows should stay in Claude Code vs. another provider

This gives you useful data before the July cutoff.

Where API work fits in

For backend and platform engineers, API work is one of the highest-leverage uses of the extra quota.

Claude Code is strong at:

• Writing route handlers
• Generating OpenAPI specs
• Debugging request/response mismatches
• Creating integration tests
• Updating generated clients
• Refactoring validation logic

These tasks often require deep context and frequent tool calls, which makes them quota-heavy.

A practical contract-first workflow looks like this:

1. Design the API contract in Apidog.
   
   Define endpoints, request schemas, response schemas, and example payloads.

2. Export the OpenAPI spec.
   
   Use it as the source of truth for Claude Code.

3. Give Claude Code an implementation goal.
   

   /goal Implement the billing API according to openapi.yaml.
   Success criteria:
   - All endpoints in the spec are implemented
   - Request validation matches the schema
   - Response payloads match examples
   - Existing tests pass
   - Apidog test cases pass

1. Run Apidog CLI tests as the validator.
   
   This keeps Claude Code aligned with the real contract instead of an invented one.

2. Iterate until tests pass.
   
   With the higher Claude Code limits, this loop can run longer without being split across multiple sessions.

For a deeper walkthrough, see the design-first API workflow guide.

If you have not used Apidog before, download Apidog and try a contract-first workflow with the extra Claude Code quota.

What about free Claude API access?

If you do not want to pay for Claude Code, the free Claude API access guide covers available paths from Anthropic and partners.

Those options are separate from Claude Code paid-plan quotas. The 50% weekly limit increase applies only to Claude Code plans:

• Pro
• Max
• Team
• Seat-based Enterprise

It does not apply to direct Anthropic API usage.

What this does not change

This update does not:

• Change Claude API rate limits for direct API users
• Change pricing on any Claude Code tier
• Add new Claude Code features
• Change billing for enterprise seat-based plans

It is a usage cap increase. The benefit depends on how effectively you use the extra room.

FAQ

When does the 50% increase end?

July 13, 2026 at 6PM PDT, which is 1AM GMT on July 14.

Anthropic has not announced what happens after that. Assume the limit returns to baseline unless Anthropic says otherwise.

Do I need to enable anything?

No. The limits are already raised on your account.

Check current usage with:

/usage

Does the 50% increase apply to the 5-hour limit too?

No. The 50% increase applies to the weekly limit.

The 5-hour limit was separately doubled the week before. Both increases are active through July 13.

What happens if I upgrade plans during the window?

Anthropic has not published specific guidance for this case. Based on how usage windows have historically worked, the new tier’s raised limits should apply from the moment you upgrade.

Does this affect Claude Code on Anthropic API plans?

No. This change applies to Claude Code plans only.

If you call Claude directly through the Anthropic API, your API rate limits are governed separately and have not changed.

Will the higher limit become permanent?

Unknown. Anthropic framed the increase as temporary through July 13.

Use the window to test your real workload and decide whether your current plan still fits.