Skip to content

Allow context window env override for all builds #31

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
elliothux wants to merge 2 commits into
base: main
Choose a base branch
from
+7 −8
Open

Allow context window env override for all builds #31

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
01b17eb
Allow context window env override for all builds
elliothux Apr 16, 2026
6531db9
refactor: align error handling for context window limit
elliothux Apr 16, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
8 changes: 5 additions & 3 deletions src/services/api/claude.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2281,13 +2281,15 @@ async function* queryModel(
max_tokens: maxOutputTokens,
output_tokens: usage.output_tokens,
})
// Reuse the max_output_tokens recovery path — from the model's
// perspective, both mean "response was cut off, continue from
// where you left off."
// Keep the public error shape aligned with max_output_tokens so
// existing clients keep the same UX, and stash the precise
// overflow kind in errorDetails for the query loop's recovery
// routing.
yield createAssistantAPIErrorMessage({
content: `${API_ERROR_MESSAGE_PREFIX}: The model has reached its context window limit.`,
apiError: 'max_output_tokens',
error: 'max_output_tokens',
errorDetails: 'context_window_exceeded',
})
}
break
Expand Down
7 changes: 2 additions & 5 deletions src/utils/context.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -52,14 +52,11 @@ export function getContextWindowForModel(
model: string,
betas?: string[],
): number {
// Allow override via environment variable (ant-only)
// Allow override via environment variable.
// This takes precedence over all other context window resolution, including 1M detection,
// so users can cap the effective context window for local decisions (auto-compact, etc.)
// while still using a 1M-capable endpoint.
if (
process.env.USER_TYPE === 'ant' &&
process.env.CLAUDE_CODE_MAX_CONTEXT_TOKENS
) {
if (process.env.CLAUDE_CODE_MAX_CONTEXT_TOKENS) {
const override = parseInt(process.env.CLAUDE_CODE_MAX_CONTEXT_TOKENS, 10)
if (!isNaN(override) && override > 0) {
return override
Expand Down