skyllwt

skyllwt/AutoSci

Skill Claude CodeCodex

Ask the wiki a question, retrieve and synthesize relevant pages, optionally crystallize the answer back into the wiki.

skyllwt/AutoSci

Skill Claude CodeCodex

Scan the full wiki to detect health issues and produce a tiered fix-recommendation report (covers all entity types in runtime/schema/entities.yaml + graph consistency).

skyllwt/AutoSci

Skill Claude CodeCodex

Run or manage the daily arXiv recommendation feed. Use for one-off fresh-paper recommendations, scheduled GitHub Actions setup/status/disable, email digests, and explicit high-confidence auto-ingest through /ingest.

skyllwt/AutoSci

Skill Claude CodeCodex

Build a ranked shortlist of candidate papers (anchor-driven, topic-driven, or derived from current wiki state) that the user — or an upstream skill — may decide to feed into /ingest. Use whenever the user asks "what should I read next", "find papers similar to this one", "recommend related work", "what's around this…

skyllwt/AutoSci

Skill Claude CodeCodex

Add or remove raw sources, or update wiki content, per user request.

skyllwt/AutoSci

Skill Claude CodeCodex

Idea-driven experiment design with iterative ablation — method candidate generation (direct / hybrid / cross-idea combination) → benchmark selection → iterative ablation (non-linear: ablation can trigger method simplification and re-planning) → sensitivity analysis → main experiment → optional generalization → deep…

skyllwt/AutoSci

Skill Claude CodeCodex

Experiment verdict gate — Review LLM independently judges results → 4 verdict paths → auto-update the linked idea's status / failurereason and graph edges.

skyllwt/AutoSci

Skill Claude CodeCodex

Pilot result evaluation — read pilot results, apply success criteria, update idea page (pilotresult, failurereason if failed), generate PILOTVERDICTREPORT. Called by /ideate Phase 5 after /exp-pilot-run.

skyllwt/AutoSci

Skill Claude CodeCodex

Pilot experiment execution — read Pilot Spec YAML, write pilot code, run experiment(Confirm with the user before operation and require the applicant to conduct manual inspection), return results. Called by /ideate Phase 5. Does NOT modify wiki pages or judge pass/fail.

skyllwt/AutoSci

Skill Claude CodeCodex

Full experiment execution pipeline — prepare code → deploy(Confirm with the user before operation and ask the applicant to conduct manual inspection) → monitor → collect results, supporting three run modes.

skyllwt/AutoSci

Skill Claude CodeCodex

View the status of all running experiments; optionally auto-collect completed experiments and advance the pipeline.

skyllwt/AutoSci

Skill Claude CodeCodex

Multi-phase research idea generation pipeline: landscape scan → dual-model brainstorm → filter & validation → write to wiki → pilot.

skyllwt/AutoSci

Skill Claude CodeCodex

Ingest a paper into the wiki — creates pages (papers + concepts + methods + people) and builds all cross-references and graph edges. Trigger whenever the user says "ingest", "add this paper", drops a .pdf / .tex / arXiv URL, or asks to fold a paper into the knowledge base.

skyllwt/AutoSci

Skill Claude CodeCodex

Bootstrap ΩmegaWiki from user sources plus optional discovery, then ingest the final paper set in parallel.

skyllwt/AutoSci

Skill Claude CodeCodex

Multi-source novelty verification — WebSearch + Semantic Scholar + wiki + Review LLM cross-verify — outputs novelty score and recommendations. Optionally writes the score back to an idea page with --write.

skyllwt/AutoSci

Skill Claude CodeCodex

LaTeX compile → PDF — latexmk compile + auto-fix + page count/anonymity/font/[UNCONFIRMED] checks + submission checklist.

skyllwt/AutoSci

Skill Claude CodeCodex

Draft a LaTeX paper from PAPERPLAN — write each section from wiki sources + generate figures/tables + BibTeX verification + de-AI polish.

skyllwt/AutoSci

Skill Claude CodeCodex

Compile a paper outline from the idea graph — evidence map → narrative structure → section plan + figure plan + citation plan, Review LLM review mandatory.

skyllwt/AutoSci

Skill Claude CodeCodex

Generate an academic poster from a drafted paper — distill sections into a single-page HTML poster with figures and inter-section transitions.

skyllwt/AutoSci

Skill Claude CodeCodex

Seed wiki/foundations/ with domain background knowledge so subsequent /ingest does not create duplicate concept pages for textbook material.

skyllwt/AutoSci

Skill Claude CodeCodex

Parse review comments → atomize concerns (Rvx-Cy) → map to wiki ideas/methods → check evidence → Review LLM stress-test → generate rebuttal.

skyllwt/AutoSci

Skill Claude CodeCodex

General-purpose multi-round iterative improvement — repeatedly calls /review on any research artifact, parses feedback, applies fixes, updates wiki, until the target score is reached.

skyllwt/AutoSci

Skill Claude CodeCodex

End-to-end research orchestrator — idea discovery → experiment design → execution → verdict → paper writing, with human gates and session-resumable state.

skyllwt/AutoSci

Skill Claude CodeCodex

Reset wiki state to a clean scaffold by scope (wiki / raw / log / checkpoints / all). Useful during development or carefree restarts after a botched setup.