Skip to content

fix(cli): recover from provider context-limit errors by compacting #14635

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

Merged
eshurakov merged 2 commits into from
Sep 28, 2026
Merged

fix(cli): recover from provider context-limit errors by compacting #14635

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
0faccb0
fix(cli): recover from provider context-limit errors by compacting
eshurakov Sep 28, 2026
55bd657
refactor(llm): match context-length overflow without number format
eshurakov Sep 28, 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
5 changes: 5 additions & 0 deletions .changeset/context-overflow-recovery.md
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,5 @@
---
"@kilocode/cli": patch
---

Recover from provider context-limit errors by compacting and retrying when the provider reports the request exceeded the model's context length.
4 changes: 4 additions & 0 deletions packages/llm/src/provider-error.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -29,6 +29,10 @@ const patterns = [
/model_context_window_exceeded/i,
/too many tokens/i,
/token limit exceeded/i,
// kilocode_change start - providers/gateways report over-long requests as
// "<N> tokens long and exceeds this model's context length of <M> tokens"
/exceeds (?:this|the) model'?s (?:(?:maximum|max) )?context length(?: of [\d,]+ tokens?)?/i,

@WebReflection WebReflection Sep 28, 2026

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

it's all good but I wonder if that translates well in terms of i18n ... the , for thousands is not universally common so that:

  • why do we have a thousand separator at all?
  • once we answer previous point, should we add all other thousand separators in there too?

not a blocker, rather a question, and once answered, I'll approve

// kilocode_change end
]

// kilocode_change start - keep transient token throttles retryable
Expand Down
4 changes: 4 additions & 0 deletions packages/llm/test/provider-error.test.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -13,6 +13,10 @@ describe("provider error classification", () => {
"Prompt has 5,958,968 tokens, but the configured context size is 256,000 tokens",
"Too many tokens",
"Token limit exceeded",
// kilocode_change start - gateway/provider wording seen in production
"The request is 280913 tokens long and exceeds this model's context length of 262144 tokens.",
"[Nex AGI] The request is 282364 tokens long and exceeds this model's context length of 262144 tokens.",
// kilocode_change end
]

expect(messages.every(isContextOverflow)).toBe(true)
Expand Down
Loading