feat: add context_tokens to internal Agents State - #12102
Merged
Conversation
|
The latest updates on your projects. Learn more about Vercel for GitHub. 1 Skipped Deployment
|
Contributor
Coverage report
Click to see where and how coverage changed
This report was generated by python-coverage-comment-action |
||||||||||||||||||||||||||||||
julian-risch
approved these changes
Jul 23, 2026
julian-risch
left a comment
julian-risch
left a comment
Member
There was a problem hiding this comment.
Looks very good to me! 🚀
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Related Issues
Proposed Changes:
The
Agentnow tracks an approximate current context-window size in its internalStateundercontext_tokens, refreshed after every LLM call with that reply's prompt-plus-completion tokens (normalized across theprompt_tokens/completion_tokensandinput_tokens/output_tokenskey conventions). Unliketoken_usage, which accumulates across the whole run,context_tokensis replaced each call. Hooks can read it viastate.get("context_tokens")— for example, abefore_llmhook that triggers context compaction once the value crosses a threshold. It is a best-effort snapshot: it is0when the generator does not report usage, and does not count messages appended after the latest call until the next call refreshes it.How did you test it?
New tests
Notes for the reviewer
I decided to add a new utils file
agent/utils.pyto start collecting these util functions that are crowding the top of theagent.pyfile. After this PR I will open a new one to move more util functionality into that file. This is not a breaking change since all util functions being moved are private.I also ran live runs against different model providers APIs using our chat generators to see what usage looks like and confirm that the agent's util methods works for our most popular providers. I've collected them in
test_normalizes_provider_shapesChecklist
fix:,feat:,build:,chore:,ci:,docs:,style:,refactor:,perf:,test:and added!in case the PR includes breaking changes.