Light0305/Light-skills

Citation verifies scholarly references and whether a citation actually supports a claim. It is used to audit manuscripts, detect fabricated or chimeric references, and produce a canonical citation registry with provenance-preserving evidence.

Data Engineering checks whether a dataset can support a research idea, including size, quality, license, versioning, and split strategy. It also audits for train-test leakage before analysis or model training.

Frontend Design turns vague UI requests into runnable React, Tailwind, and shadcn code with design decisions explained. Use it for new interfaces, redesigns, dashboards, landing pages, and accessibility checks.

Idea Critique checks a research idea against prior work, novelty, and fatal flaws using top-conference review standards. Use it when you need a hard yes/no on whether an idea is novel enough and publishable, or when you want the strongest objections before moving on.

Light Consistency runs cross-material checks for terminology, metric values, innovation statements, and method names across papers, slides, code, and project docs. Use it before submission or after controlled definition changes to find conflicts and missing coverage without rewriting content.

Light Experiment Coding implements frozen research plans as minimal, test-first experiment code with controls for leakage and reproducibility. Use it to build or review training, preprocessing, and evaluation runs with raw artifacts for result analysis.

Light Figure plans and renders publication-ready scientific figures that support specific claims. It is used when turning analysis results into paper figures, checking visual honesty, and reviewing rendered charts for layout issues.

Light File Reading deeply understands user-provided files like PDFs, Word docs, spreadsheets, slides, images, code, and archives, then produces structured notes and next-step mappings. Use it when you need to know what a file says, how it is organized, or what can be done with it next.

Light Idea Generation structures research ideation from a direction, dataset, or literature map into layered candidate ideas. It is used when you need research questions, innovation angles, or a first pass before critique and feasibility review.

Light Memory PM keeps a project's status, decisions, version history, terminology, and handoff notes in per-project .light/ files. Use it for long-running work that needs cross-session continuity, resume summaries, or controlled terminology changes.

Light Orchestrator coordinates multi-stage Light research projects through passport state, checkpoints, handoffs, and recovery. Use it to resume, reroute, or deliver work across stages 1–13 when a project is new, partial, dirty, failed, or stale.

Light Review Rebuttal builds auditable review, revision, and author-response packages from reviewer comments and venue context. It is used when drafting rebuttals, triaging revisions, or simulating a pre-submission review.

Light System Design designs or modernizes software systems from current-state inventory through architecture options, contracts, migration and rollback plans, ADRs, and verification. Use it for greenfield systems, monolith modernization, API or schema changes, and reliability or tenant/PII controls.

Light Typesetting builds and preflights submission-ready LaTeX/PDF artifacts for stage 11. Use it to choose an engine and bibliography backend, diagnose compile errors, and check page limits, anonymity, metadata, fonts, figures, tables, and citations.

Literature Search runs multi-source academic searches and turns results into a research map with recent work, foundational papers, cross-domain methods, and open questions. Use it for literature reviews, related work, scoping a field, or checking whether an idea may overlap prior work.

Paper Writing helps draft and revise papers around evidence-backed claims, matching wording strength to evidence and checking consistency across abstract, introduction, and conclusion. Use it when writing or polishing a manuscript, especially to avoid overclaiming or unsupported claims.

Patent Disclosure Handoff turns a project or research result into an attorney-reviewable invention disclosure packet with evidence, figures, claim support, and open questions. Use it for patent-point mining, prior-art context, or drafting a technical disclosure for counsel review.

Project Structure Lifecycle audits, plans, scaffolds, and safely migrates research project trees across greenfield, existing, and monorepo roots. Use it for source inventories, move maps, naming policy, profile selection, and rollback while preserving uncommitted work and .light content.

Research Ethics checks research outputs for integrity, consent, disclosure, and publication-risk issues. Use it before submission, data or code release, and human- or animal-subject research decisions.

Research Plan turns a vetted idea into an executable research plan with question, estimand, experiment matrix, power checks, and preregistration fields. Use it when designing experiments, ablations, baselines, or reproducible study protocols.

Result Analysis interprets experiment outputs with significance tests, effect sizes, confidence intervals, and multiple-comparison correction. It is used when results need to be tied to claims, evidence strength, and reproducibility without p-hacking.

Software Copyright Materials prepares draft China software copyright registration materials from a real project, including a worksheet, source-code deposit plan, user manual, screenshots, and consistency checks. It is used for 软著 pre-submit review and local formal-output evidence, not legal advice or filing.

Venue Matching builds evidence-bound journal or conference shortlists for a paper after typesetting delivers the final handoff. It is used to match scope, article type, reach, OA/APC, indexing, deadline, and venue risk before a user makes the final choice.