Long Description Handling

Description

--- ⚠️ AI-Generated Skill Generated by MFKVault on 2026-05-14. Review before use. Not professional advice. Modify as needed for your use case. --- --- name: Long Description Handling description: Process and work with text descriptions that exceed twenty characters, breaking them into manageable chunks and extracting key information systematically. --- # Long Description Handling ## When to use this skill Use this skill when you receive verbose descriptions, lengthy specifications, or detailed narratives that need to be parsed, summarized, or acted upon. This applies to documentation, requirements, user stories, feature descriptions, or any multi-sentence content that requires systematic analysis rather than quick pattern matching. ## Key behaviors - **Segment long text into logical units**: Break descriptions at natural boundaries (sentences, paragraphs, sections) to process information in digestible chunks rather than treating the entire block as one unit. - **Identify and extract key entities**: Pull out nouns, named items, technical terms, and proper nouns that represent the core subjects being discussed. - **Recognize hierarchical structure**: Detect primary concepts versus supporting details, main objectives versus constraints, and prerequisites versus outcomes. - **Preserve context across segments**: Maintain reference to earlier parts of the description when analyzing later sections to ensure coherent understanding. - **Generate accurate summaries**: Condense long descriptions into concise bullet points or single-sentence summaries that capture the essential meaning without losing critical details. - **Ask clarifying questions strategically**: When descriptions are ambiguous, vague, or contradictory, request specific information about the missing or unclear portions rather than making assumptions. - **Map descriptions to actionable tasks**: Convert narrative descriptions into concrete, enumerated steps or requirements that can be implemented or followed. ## Examples ### Example 1: Long Feature Description **User request**: "I need you to help me build a system that allows users to upload images, apply various filters like sepia, black and white, and blur effects, store the results in a cloud database, retrieve them later, and share filtered images with other users through a simple link-sharing mechanism." **Response approach**: 1. Break into components: image upload → filter application → storage → retrieval → sharing 2. Extract key entities: images, filters (sepia, black and white, blur), cloud database, link-sharing 3. Identify requirements: upload capability, filter suite, cloud storage, retrieval system, sharing links 4. Summarize: "Build an image filter application with cloud storage and link-based sharing" 5. Ask clarifying question: "Should users be able to chain multiple filters, or apply one at a time?" ### Example 2: Lengthy Technical Specification **User request**: "Our application needs to validate user input from multiple sources including form fields, API requests, CSV uploads, and database imports. The validation rules vary by source—form fields should show inline error messages, API requests should return specific error codes, CSV uploads should generate detailed error reports, and database imports should log validation failures. All validation must happen in real-time for forms and API, but batch validation is acceptable for imports." **Response approach**: 1. Identify validation sources: forms, APIs, CSV, database 2. Map validation rules by source: - Forms: inline errors - APIs: error codes - CSV: detailed reports - Database: logged failures 3. Distinguish timing requirements: real-time (forms/APIs) vs. batch (imports) 4. Extract core requirement: "Build multi-source input validation with source-specific error handling and differentiated timing" 5. Propose structure: separate validators per source with unified validation logic ### Example 3: Complex User Story **User request**: "As a project manager, I want to be able to view all tasks assigned to my team members across different projects, filter them by priority and due date, see which tasks are blocked by dependencies, identify team members with excessive workload, and generate a weekly report showing task completion rates, so that I can better allocate resources and ensure on-time delivery." **Response approach**: 1. Extract actor: project manager 2. Identify core needs: task visibility, filtering, dependency analysis, workload assessment, reporting 3. Break into features: - Aggregated task view (cross-project) - Filtering (priority, due date) - Dependency visualization - Workload metrics - Automated reporting 4. Recognize underlying goal: resource optimization and delivery management 5. Suggest structure: dashboard with multiple views and automated weekly reports ## What NOT to do - **Don't truncate descriptions without summarizing**: If a description is long, don't simply ignore parts of it or assume you understand without reading the full context. - **Don't treat long descriptions as single monolithic requirements**: Avoid attempting to implement everything at once; always decompose into smaller, manageable pieces. - **Don't make assumptions about unclear details in lengthy text**: When descriptions contain ambiguous language, ask for clarification rather than guessing the user's intent. - **Don't lose sight of the primary goal**: While breaking down details, maintain focus on the overarching objective stated in the description. - **Don't skip over constraints and edge cases**: Long descriptions often contain important limitations or special conditions buried in middle paragraphs—flag these explicitly. - **Don't respond with equally long rambling text**: Keep responses focused and organized, mirroring the structured approach you use to parse input. ## Edge cases How to handle: - **Circular or contradictory requirements**: When a long description contains conflicting statements (e.g., "must be real-time" and "batch processing is acceptable"), ask the user to prioritize or clarify which applies in which scenario. - **Missing context from incomplete descriptions**: If a long description references undefined terms or prior context you lack, explicitly state what information is missing and ask for it before proceeding. - **Extremely verbose descriptions with minimal actionable content**: For descriptions that ramble or repeat concepts without adding detail, politely summarize what you understand and ask if it's accurate, then propose a more concise structure. - **Multi-format descriptions mixing narrative and technical specs**: When descriptions blend natural language explanations with code examples, technical constraints, and user stories, organize your response with separate sections for each format. - **Descriptions with implicit requirements**: Long descriptions sometimes imply unstated requirements (e.g., "generate reports" implies data aggregation, permissions, scheduling). Identify and surface these implicit needs for confirmation.

Security Status