Repository navigation
feat(ai): Classify LLM cache usage - #127015
Merged
Merged
Conversation
vgrozdanic
commented
Oct 9, 2026
Comment on lines
+16
to
+18
| # https://docs.anthropic.com/en/docs/build-with-claude/prompt-caching | ||
| # https://platform.openai.com/docs/guides/prompt-caching | ||
| # https://ai.google.dev/gemini-api/docs/caching |
Member
Author
There was a problem hiding this comment.
in order to keep this up to date, we'll set up Junior periodic task for now, there's no other way of having this information at the moment
vgrozdanic
marked this pull request as ready for review
October 9, 2026 11:52
ArthurKnaus
approved these changes
Oct 12, 2026
vgrozdanic
force-pushed
the
feat/llm-cache-classification
branch
from
October 12, 2026 15:28
92b310e to
6db179a
Compare
This was referenced Oct 15, 2026
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Adds provisional classification for the LLM prompt-cache shadow detector. Call sites with enough traffic and stored evidence are classified as healthy, not caching, or repeatedly writing cache entries without reading them back.
Classification uses cache hit rate, cache creation volume, prompt size, call volume, and stored-span count. All models use a 4,096-token prompt threshold because savings below that size are not material enough to report. This avoids model-specific policy and maintenance.
These results remain provisional. This layer does not query telemetry, schedule detection, emit logs, or create issues. Later stack layers validate traffic spacing and instrumentation before treating a classification as a shadow finding.