ワンクリックで
LuminaCode
LuminaCode には GongShichen から収集した 47 個の skills があり、リポジトリ単位の職業カバレッジとサイト内 skill 詳細ページを表示します。
このリポジトリの skills
Convert source notes into claim-level evidence with confidence and limitations.
Verify source traceability, artifact completeness, and citation correctness before citation_qa gate.
Write an evidence-backed final research report from sources and the evidence matrix.
Review research method quality, source diversity, bias, and conclusion strength.
Convert a broad research request into subquestions, assumptions, and source criteria.
Design and execute a multi-round SearxNG/arXiv search strategy.
Read a web page or paper and produce source-grounded notes with stable IDs.
Define a durable DeepResearch acceptance contract before dispatching research work.
Prepare backend implementation evidence and integration handoff for frontend, QA, and Reviewer.
Define service, API, event, storage, or process-boundary contracts between components.
Plan durable state, data model, migrations, caching, and recovery behavior.
Design runtime ownership, concurrency, async work, and isolation for the product's backend or platform layer.
Plan benchmark, evaluation, load-test, or experiment environment and reporting behavior.
Verify build, install, resource copy, and uninstall behavior.
Verify release readiness after implementation.
Keep generated runtime, verification, cache, log, and build artifacts inside the intended project root.
Verify and document how the user-facing surface consumes the agreed backend/API contract.
Plan user-facing state, rendering, interaction, and integration changes for any interface stack.
Design polished text-first and terminal interfaces with stable layout and readable interaction states.
Review command-line or terminal interaction behavior and state transitions.
Provide concrete verification commands and manual acceptance steps.
Write a durable QA report with evidence, command results, unresolved findings, and verdict rationale.
Identify likely regressions from a proposed change.
Convert requirements into focused unit, integration, and manual tests.
Identify benchmark, evaluation, experiment, or comparison requirements and upstream constraints.
Produce concise domain context and constraints without taking over product or technical ownership.
Produce a concise evidence-backed research report.
Map repository structure, ownership boundaries, and relevant files.
Review system boundaries, coupling, and long-term risks.
Review implementation changes for correctness and maintainability.
Review whether the product-development process and required artifacts were actually followed.
Review permissions, local IPC, filesystem, and command execution risks.
Check that each product-development stage has durable documents before moving to the next gate.
Verify Team completion policy before calling CompleteTeamTask.
Merge specialist outputs into a coherent plan, decision, or final answer.
Produce a product requirements document that can drive design, engineering, QA, and review.
Build and maintain a team task graph with owners, dependencies, artifacts, and gates.
Structure visible dialogue, activity, history, and operational information.
Convert a PRD into a durable UX design document with states, flows, copy, and accessibility.
Write concise terminal UI copy for states, prompts, and confirmations.