What problem does it solve? Long-running research quests often end ambiguously, with unclear claim status, lost negative results, and no clean stopping point. This Skill closes or pauses a quest responsibly by producing an accurate final state: what is supported, what failed, what remains open, and whether to stop, archive, publish, or continue later. ## Core Features & Use Cases - Claim Ledger Construction: Classifies every major outcome as supported, partially supported, unsupported, or deferred, with evidence paths, caveats, and a belief-change log preserving downgrade history. - Closure Gating: Blocks premature finalization when evidence gaps, unresolved outline items, or unmapped analysis remain, routing back through the decision stage instead of forcing closure. - Resume and Handoff Packets: Produces compact restart packets naming the accepted baseline, strongest evidence, top blockers, and do-not-repeat notes for future continuation. - Use Case: After a multi-week experiment campaign with a paper bundle, use this Skill to verify the bundle manifest, refresh SUMMARY.md and status.md, record the final stop-or-publish decision, and leave a recoverable checkpoint. ## Quick Start Ask the agent to finalize the current quest by consolidating the accepted evidence, building the claim ledger, and recording a final stop, archive, or continue-later recommendation.
