fix: move dynamic ACP injections to suffix message for prefix cache stability#6
Merged
Merged
Conversation
…tability Dynamic metadata (context usage, visible IDs, block guidance, anchored nudges) was injected into the last user message on every transform call. This invalidated OpenAI Responses prefix cache for all subsequent content. Move all dynamic injections to a synthetic suffix message at the end of the message list so historical messages remain byte-stable across turns. Fixes: #5 Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-opencode) Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
Context usage and visible ID range are now injected into a suffix message at the end of the message list instead of the last user message. Update test assertions to check suffix message and account for +1 message count. Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-opencode) Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-opencode) Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-opencode) Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
…e-dcp to opencode-acp Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-opencode) Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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
Fixes #5
Problem
ACP injected dynamic per-turn metadata into the last user message on every chat.messages.transform call. Because OpenAI Responses uses prefix-based caching, modifying an early user message invalidated the cache for all subsequent content. Cache read tokens would plateau at ~25.6K while total prompt tokens grew from 38K to 83K.
Fix
Before: ACP modifies early user message -> cache breaks for all subsequent content
After: ACP appends synthetic message at end -> historical messages stay byte-stable
Files Changed
Test Results