“refinement”的搜索结果

将软件开发划分为需求、设计、测试先行的实现、审查和部署等阶段，通过Claude Flow协调多个智能体分工，管理开发全过程。

• 从需求出发开发新功能
• 结合测试改造现有代码
• 排查故障并添加回归测试

Turn descriptions of scientific components and relationships into PNG diagram drafts with Nano Banana 2. Gemini 3.7 Flash reviews each draft and may request one revision, while saved image versions and review logs support manual checking.

中文简介正在准备中。当前显示英语简介。

• Drafting neural network architecture diagrams
• Illustrating biological pathways
• Creating study flowcharts

Combine laboratory animal welfare measurements into reference-relative RELSA severity scores. Analyze individual trajectories, explore forecasts for specified observation times, and derive candidate score zones, with explicit checks on normalization, reference selection, prediction intervals, and evaluation limits.

中文简介正在准备中。当前显示英语简介。

• Combine animal welfare measurements into daily scores
• Compare group severity using a shared reference
• Explore severity forecasts at specified observation times

Organize repeated experiments to improve models, prompts, or agent systems against measurable goals. The skill records hypotheses and lessons in a persistent tree, isolates trials, and compares promising candidates using evaluation data kept separate from development.

中文简介正在准备中。当前显示英语简介。

• Compare repeated experiments to improve model training
• Improve an agent system's measured success rate
• Test prompt changes against a fixed evaluator

Clarify a task through interactive questions before carrying it out. Using Joyride’s input tool, establish scope, deliverables, constraints, and success criteria, then present a plan and create a todo list before starting work.

中文简介正在准备中。当前显示英语简介。

• Clarify the scope of a vague request
• Agree on deliverables and success criteria
• Identify technical and procedural constraints

Refine task prompts through questions about scope, deliverables, constraints, and success criteria. Receive an organized Markdown prompt in chat and on your clipboard, with revisions supported through the Joyride extension.

中文简介正在准备中。当前显示英语简介。

• Turn a vague request into a detailed task prompt
• Clarify deliverables and success criteria
• Organize technical and procedural requirements

Design workflows that evaluate and refine AI outputs through self-review, separate generation and evaluation roles, tests, or weighted scoring criteria. Includes patterns for structured feedback, iteration limits, and stopping when quality stops improving.

中文简介正在准备中。当前显示英语简介。

• Adding self-review and revision loops to AI outputs
• Separating generation from evaluation
• Refining generated code using test results

Provides prompt templates for coding, writing, analysis, and creative work. Helps define the AI's role, specify the desired output, add examples and constraints, and refine requests using a practical checklist.

中文简介正在准备中。当前显示英语简介。

• Draft a prompt for code review
• Create requests for clear technical documentation
• Prepare prompts for feature and naming ideas

Reviews pull requests for bugs, error handling, code quality, and style consistency. Organizes feedback into an overview, must-fix issues with file locations, improvement suggestions, and positive notes so reviewers can distinguish urgent problems from refinements.

中文简介正在准备中。当前显示英语简介。

• Finding bugs that must be fixed before merging
• Reviewing TypeScript types and asynchronous code
• Checking naming, imports, and leftover debug statements

Turn a vague intention into a concrete goal with clear scope, measurable success criteria, and evidence of completion. Check existing goals before creating one, so the agent has a specific outcome to pursue without duplicating active work.

中文简介正在准备中。当前显示英语简介。

• Turn a vague intention into a measurable goal
• Define completion criteria for a bug fix
• Set performance targets and measurement methods

Refine a working feature through browser feedback from the user. Collect observations through typed comments or live speech and drawing, apply focused usability fixes, and save the changes in local commits.

中文简介正在准备中。当前显示英语简介。

• Refining usability before release
• Adjusting a page while checking the changes
• Giving feedback through speech and on-page drawing

Explore one problem through several distinct perspectives to uncover hidden requirements and edge cases. Turn the findings into testable success criteria, refinements, and failure conditions, stopping when additional perspectives no longer produce new requirements.

中文简介正在准备中。当前显示英语简介。

• Finding overlooked requirements before starting work
• Reviewing a problem from stakeholders' perspectives
• Exploring failure modes and changes over time

Create model architecture diagrams and method illustrations for academic papers using Gemini. Codex plans the figure and reviews generated images for correct connections, content, layout, and readability, guiding repeated revisions toward a clearer final figure.

中文简介正在准备中。当前显示英语简介。

• Illustrate a model architecture for a paper
• Show a method's processing steps and data flow
• Create a conceptual comparison diagram

Turn a vague research direction into a focused proposal and a detailed experiment roadmap. This workflow connects method refinement with validation planning, producing review records, an experiment tracker, and a summary of what to run next.

中文简介正在准备中。当前显示英语简介。

• Turn a rough research idea into a focused proposal
• Plan experiments that test the paper's claims
• Organize experiment order, budget, and decision points

Create academic architecture diagrams and method illustrations with Gemini image generation. The workflow refines layout and visual style, then uses Claude to inspect arrows, labels, and readability and guide further revisions.

中文简介正在准备中。当前显示英语简介。

• Illustrating a model architecture for a paper
• Turning a method description into a workflow diagram
• Reviewing diagram arrows and labels

Turn method notes or figure descriptions into academic architecture and concept illustrations. Claude plans and reviews the layout, while a local Codex bridge generates images, with refinement rounds and final PNG, LaTeX snippet, and review records.

中文简介正在准备中。当前显示英语简介。

• Illustrating a model architecture for a paper
• Turning method notes into a workflow diagram
• Creating concept or comparison figures

Uses a completed ZKsync Era investigation with no reportable findings as a defense study. It explains 25 tested attack vectors, the safeguards that blocked them, and criteria for deciding whether to continue reviewing a large bridge protocol.

中文简介正在准备中。当前显示英语简介。

• Learning defensive design from a completed ZKsync Era investigation
• Reviewing why 25 tested attack vectors failed
• Assessing whether to continue investigating a large bridge

Translate user needs into concise stories that explain who benefits, what they want, and why. Add Gherkin acceptance criteria—preconditions, an action, and an expected result—to support development discussions and testable definitions of success.

中文简介正在准备中。当前显示英语简介。

• Turning requirements into user stories
• Defining testable acceptance criteria before development
• Preparing stories for backlog refinement and sprint planning

Execute an existing Ouroboros Seed, a specification containing a goal and acceptance criteria, as a background job. The workflow checks each criterion, normally queues formal evaluation, and follows a bounded improvement loop if evaluation rejects the result.

中文简介正在准备中。当前显示英语简介。

• Executing an existing Seed specification
• Checking whether each acceptance criterion is met
• Following execution, evaluation, and improvement results

Review existing interface animations and tune their timing, movement, scale, blur, and easing—the way speed changes. It matches values to their purpose and checks opening, closing, hover behavior, and staggered sequences against shared motion tokens.

中文简介正在准备中。当前显示英语简介。

• Reviewing existing interface animations
• Adjusting modal opening and closing speeds
• Fixing delays in staggered list animations

Organize software development from requirements and architecture through testing, review, and deployment. Use specialized agent roles, shared memory, and test-driven development, where tests are written before code, to coordinate work across the project.

中文简介正在准备中。当前显示英语简介。

• Plan and implement a new feature
• Refactor legacy code with supporting tests
• Review code security and performance

Turn a rough request into a structured AI prompt with a clear goal, output format, and constraints. For complex work, add smaller tasks, separate creator and reviewer roles, and explicit checks for accuracy and uncertainty.

中文简介正在准备中。当前显示英语简介。

• Refining a vague AI request
• Designing prompts for API documentation
• Adding verification steps to code generation

Turn written descriptions into scientific diagrams, including neural network architectures, biological pathways, and research flowcharts. Nano Banana 2 generates images, while Gemini 3.1 Pro Preview reviews them and guides regeneration when needed.

中文简介正在准备中。当前显示英语简介。

• Illustrating neural network architectures
• Creating participant flowcharts for research
• Drawing biological signaling pathways

Build VeADK Agent code from functional requirements, including agent structure and prompt refinement. Convert existing Langchain or Langgraph code, or a Dify workflow, into VeADK Agent code and save the generated Python files.

中文简介正在准备中。当前显示英语简介。

• Generating a VeADK Agent from functional requirements
• Converting Langchain code to VeADK Agent code
• Converting Langgraph code to VeADK Agent code