Repository navigation
Apply research-backed prompt framing to context injection - #12
Merged
Merged
Conversation
Based on EmotionPrompt and BCSP research, reword how memories are presented to the LLM to improve attention and adherence: - MCP server instructions: establish role and behavioral guidance instead of a flat description - Hook output headers: "prior learnings — apply these to avoid repeating past mistakes" instead of passive "recalled memories" - Antipattern framing: confidence-based language (KNOWN MISTAKE for high-severity, warning for others) with consequence framing - Atom formatting: high-confidence atoms get assertive framing, low-confidence atoms get "unverified" hedging - MCP prompt templates: add stakes/rationale sentence before procedural steps (why > what) All 1061 tests pass with no regressions. https://claude.ai/code/session_01J8rNURgnxHZkwshKBEU5Nb
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Based on EmotionPrompt and BCSP research, reword how memories are
presented to the LLM to improve attention and adherence:
instead of a flat description
repeating past mistakes" instead of passive "recalled memories"
high-severity, warning for others) with consequence framing
low-confidence atoms get "unverified" hedging
procedural steps (why > what)
All 1061 tests pass with no regressions.
https://claude.ai/code/session_01J8rNURgnxHZkwshKBEU5Nb