Rework system-prompt cache identity and /prompt-show output`#509
Merged
Conversation
The static prompt cache was keyed only on agent+model, so adding or removing file contexts silently reused stale instructions. Introduce a SHA-256 signature over all inputs that affect static prompt rendering (contexts, rules, skills, tools, environment, chat-prompt template); repo-map content is represented by presence only to avoid running `git ls-files` on every cache hit. Consolidate SHA-256 helpers into eca.digest and reuse in cache.clj and oauth.clj. Quote-safe attribute rendering replaces ad-hoc string formatting in context/rule/skill XML tags. /prompt-show is now read-only: it no longer consumes the editor-state cursor, and its output is reorganized into clearly delimited sections (Instructions / Chat) matching what the model receives. Fix a double blank line before ## MCP Resources caused by a trailing separator leaking when the Context section is absent.
zikajk
force-pushed
the
rework-system-prompt-handling
branch
from
24ab2d2 to
262963d
Compare
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
The static prompt cache was keyed only on agent+model, so adding or removing file contexts silently reused stale instructions. Introduce a SHA-256 signature over all inputs that affect static prompt rendering (contexts, rules, skills, tools, environment, chat-prompt template); repo-map content is represented by presence only to avoid running
git ls-fileson every cache hit.Consolidate SHA-256 helpers into eca.digest and reuse in cache.clj and oauth.clj. Quote-safe attribute rendering replaces ad-hoc string formatting in context/rule/skill XML tags.
/prompt-show is now read-only: it no longer consumes the editor-state cursor, and its output is reorganized into clearly delimited sections (Instructions / Chat) matching what the model receives. Fix a double blank line before ## MCP Resources caused by a trailing separator leaking when the Context section is absent.