CLAUDE_CODE_MAX_CONTEXT_TOKENS

Updated October 13, 2026 by Alex Sorokoletov

CLAUDE_CODE_MAX_CONTEXT_TOKENS is a Claude Code environment variable: Sets an explicit ceiling on the context window size used during a session. Rather than relying on the model's default maximum, this variable lets operators cap token consumption — useful for cost-conscious teams, rate-limited API plans, or environments where shorter context windows improve latency. Added between v2.1.96 and v2.1.100 and still present in v2.1.294. It is listed in the official reference.

How to set it

For one session, export it in the shell that starts Claude Code:

export CLAUDE_CODE_MAX_CONTEXT_TOKENS=<value>
claude

To apply it to every session, add it to the env object of a settings file:

// ~/.claude/settings.json
{
  "env": {
    "CLAUDE_CODE_MAX_CONTEXT_TOKENS": "<value>"
  }
}

Release notes that mention it

• Claude Code v2.1.100 — 3 New Environment Variables: Context Token Limits, Perforce VCS, and Script Capability Controls
• Claude Code Version Tracker: Undocumented Env Vars, Feature Gates & Models

Related variables

• CLAUDE_CODE_FILE_READ_MAX_OUTPUT_TOKENS: Max tokens for file read output.
• CLAUDE_CODE_MAX_OUTPUT_TOKENS: Maximum output tokens; validated per-model.
• MAX_MCP_OUTPUT_TOKENS: Max tokens for MCP tool output.
• MAX_THINKING_TOKENS: Likely controlled via API settings now.
• CLAUDE_CODE_IDLE_COMPACT_MIN_TOKENS: Sets the minimum conversation size, in tokens, before idle compaction runs.
• CLAUDE_BG_SOCKET_TOKENS_PATH
• CLAUDE_CODE_TOTAL_TOKENS_REMINDER
• CLAUDE_CODE_TOTAL_TOKENS_REMINDER_AFTER_USER_TURN

Frequently asked questions

All variables, searchable: Claude Code environment variables.