fix: simplify lever prompt to restore content quality#297
Merged
Conversation
External review rated baseline report 6.5/10 vs optimized 5.8/10: "Improved specificity, regressed in credibility." The system prompt was forcing verbose, overconfident output with fabricated numbers. Changes: - Remove mandatory Immediate→Systemic→Strategic chain format - Remove mandatory % change / cost delta quantification - Remove conservative→moderate→radical progression template - Remove "Radical option must include emerging tech/business model" - Add anti-fabrication: "Do not fabricate percentages or cost estimates" - Add anti-marketing: "NO marketing language" - Shorten target consequence length from 3-5 to 2-4 sentences - Simplify Pydantic field descriptions to match
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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
Immediate → Systemic → Strategicconsequence chain formatconservative → moderate → radicaloption progression templateMotivation
External review of full PlanExe reports (hong_kong_game) rated the baseline report 6.5/10 and the report built on optimized levers 5.8/10: "Version 2 improved specificity, but regressed in credibility."
The optimization loop improved structural compliance (97.1% success rate) but degraded content quality. Baseline consequences averaged 269 chars; iteration 17 averaged 980 chars (3.6x) with fabricated percentages and marketing-copy tone.
Test plan