Refresh docs and LLM prompts; add game-analysis skill#415
Merged
Conversation
LLM prompt fixes (fewer invalid first attempts, sharper point play): - Hand display shows pair counts; the JSON example no longer seeds a hallucinated pair; a pair is only valid when the card shows a count. - Add the general follow-suit obligation to the fixed rules. - Contribute biggest spare points (incl. K) on a locked teammate win, keeping only a live boss; align the seat-guidance bullet. - Ruff for an opponent's points; never sluff a 5/10/K into a lost trick. Docs overhaul (stop mirroring code, the main drift source, and fix errors): - README: drop stale versions, surface optional LLM players, trim grandiosity/redundancy, remove Future Enhancements. - AI_SYSTEM: add the LLM trick-play layer; remove the false DuckDB claim, invented metrics, the 2025 changelog, mermaid diagrams, the duplicated hierarchy, and exact scoring pseudocode; fix constants/overclaims. - GAME_RULES: correct the kitty multiplier (2^(pairs+1)) and shutout (+3). - MULTI_COMBO: replace stale code signatures with a concept map + pointers. - VERSIONING: fix the broken path; de-mirror the dev-client jq values. Housekeeping: - Move the log-event schema into a new game-analysis skill (wraps the repo-root analysis/ pipeline); remove the standalone doc and the stale LLM_AI_INTEGRATION proposal; add the LLM_GAME_STATE_SIGNALS future-work note. - Remove the throwaway analysis log; gitignore .vscode/. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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
2^(pairs+1), shutout+3ranks), MULTI_COMBO (stale signatures → concept map), VERSIONING (broken path, de-mirrored jq values).analysis/pipeline (analysis/left in place). Removes the standalone schema doc and the staleLLM_AI_INTEGRATIONproposal; adds theLLM_GAME_STATE_SIGNALSfuture-work note..vscode/.Guiding principle: docs hold concepts/rules/"why"; code stays the source of truth for signatures, constants, and field schemas.
🤖 Generated with Claude Code