Improve shared Stellaris game knowledge - #39
Draft
gitmaan wants to merge 7 commits into
Draft
Conversation
Centralize version-aware mechanics resolution with exact, partial, and unsupported coverage states so stale rules are not presented as current facts. Inject the same evidence hierarchy and Pegasus 4.4.6 snapshot into Advisor and every Chronicle generation path, including structured retries and recaps. Add bundle resource-path coverage, source maintenance guidance, and a provider-neutral raw-response harness for flexible human evaluation across current models.
Retry one empty provider response during structured Chronicle generation using the existing corrective-attempt budget. Route Chronicle knowledge evaluations through the production Chronicle wrapper so provider tests cover schema fallback, validation, and retry behavior end to end.
Document how version-aware mechanics snapshots support models with different training cutoffs across Advisor and Chronicle. Record the human-reviewed stale-premise test outcome and the maintenance workflow for future game releases.
Audit every stable Stellaris 4.x release line and hotfix through Pegasus 4.4.6 against official Paradox announcements, then verify disputed values against the installed game data and representative save structure.\n\nCorrect the population growth floor guidance, add missing material systems from Lyra through Pegasus, record DLC and source provenance, and extend regression plus human-review model cases. Keep Cygnus open-beta mechanics excluded from the stable snapshot.
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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
Why
Modern API and local models have different knowledge cutoffs and can confidently apply pre-4.x mechanics. This adds a compact current baseline without teaching each provider separately or treating general game rules as campaign events.
Impact
Gemini, OpenRouter, Ollama, LM Studio, and custom endpoints receive the same current mechanics context in both Advisor and Chronicle. Older or newer game versions are handled conservatively, and mods or save evidence remain authoritative.
Validation
python3 -m pytest -q(377 passed, 1 skipped)