fix(llm): preflight compact oversized requests#284
Merged
Conversation
golemcore1
force-pushed
the
fix/llm-context-preflight
branch
from
89984f6 to
4193fd3
Compare
golemcore1
force-pushed
the
fix/llm-context-preflight
branch
from
4193fd3 to
832f75f
Compare
Add targeted tests that exercise reachable defensive null-check arms: explicit messages(null) override on AgentSession builder to bypass Lombok @Builder.Default, degenerate coordinator with null orchestration service, and empty-messages preflight attempt. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
|
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
This PR fixes a long-chat failure mode where the agent could build an LLM request larger than the selected model can accept and only discover the problem after sending the request to the provider.
Changes included:
ContextTokenEstimatorfor estimating full LLM request size, including system prompt, conversation view, tools, and tool results.LlmCallPhaseafter context assembly and model selection, immediately before the provider call.ContextAttributes.LLM_REQUEST_PREFLIGHTkey.REQUEST_PREFLIGHTas a distinct compaction reason for observability.too_many_tokensas rate-limit retry signals in the LangChain4j adapter.AutoCompactionSystemso pre-context estimates are less ad hoc.