discovery
Discovery
Transform ambiguous instructions into structured requirements memos through iterative dialogue. This skill is the "Explore" phase -- understanding what needs to be done before anyone writes a plan or touches code.
Why This Exists
When work starts from a vague instruction, the natural tendency is to jump straight into solutions. This pollutes the context with misdirected work, degrades decision quality, and makes course correction expensive. Discovery forces a pause: understand first, then act.
The output is a requirements memo -- a file that captures what was learned during the dialogue. It serves as the clean handoff point to the next phase (planning, requirements definition, implementation) in a fresh session, free of the exploratory context.
How It Works
Discovery is a conversation loop with a shared artifact. The requirements memo file is created on the first turn and updated every turn. The user has the file open and reads it as the conversation progresses. This is not a one-sided interrogation -- it is a collaborative structuring of the user's thoughts, mediated by a living document.
The memo is a byproduct, not the goal. The goal is to help the user think clearly about what they need. The memo captures what emerges from that conversation. If the user is talking freely and productively, the memo will fill itself. Never steer the conversation toward empty memo sections -- steer it toward whatever the user most needs to think through.
User speaks --> Update memo file --> Respond (may include a question) --> User speaks --> repeat
^ |
More from caldiaworks/caldiaworks-marketplace
requirements-docx
Convert USDM/EARS requirements documents from Markdown to professionally formatted Word (.docx) files for client submission. Generates cover pages, table of contents, headers/footers, and styled requirement hierarchies. Leverages the docx skill for Word file generation. Use when: export requirements to Word, requirements to docx, USDM to Word, convert requirements document, 要件書をWord出力, 要件定義書のdocx変換, Word形式で要件書を作成, 要件定義書をWordに変換.
37ideation
Turn rough ideas into structured, validated idea documents through collaborative dialogue. Explores context, asks clarifying questions one at a time, proposes alternative approaches with feasibility evaluation, and produces documents ready for requirements definition. Use when: ideation, brainstorm, new idea, explore an idea, I want to build, what if we, let's think about, propose approaches, evaluate this idea, idea document, アイデア出し, 案出し, ブレスト, アイデアを整理, 検討したい.
36usdm
Convert ambiguous user requests into structured USDM requirements documents. Decomposes requirements into Requirement -> Reason -> Description -> Specification hierarchy. Integrates with GitHub Issues, Asana, and Jira tickets as input sources. Use when: create requirements, write requirements document, USDM, decompose requirements, requirements definition, 要件定義, 要件を整理, 要件分解.
34ears
Write unambiguous specifications using EARS (Easy Approach to Requirements Syntax) patterns. Provides 6 sentence templates that eliminate ambiguity: Ubiquitous, Event-driven, State-driven, Unwanted behavior, Optional feature, and Complex. Use when: EARS, specification writing, write specs, 仕様を書く, EARS記法, 仕様を明確化, requirements specification, unambiguous specification.
34skill-creator
Create new skills, improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, update or optimize an existing skill, run evals to test a skill, or benchmark skill performance with variance analysis.
30en-explainer
Explain English technical documents and text in Japanese with contextual understanding. Not a simple translator -- reads the surrounding file or codebase context to provide deeper, more accurate explanations tailored for Japanese-speaking developers. Use when: explain this English, この英文を解説, 英語の解説, en-explainer, what does this mean, この英文の意味, 英文を日本語で説明, ドキュメントを解説, README解説, エラーメッセージの意味, コメントの意味, API仕様の解説, or when the user pastes English text and asks for explanation in Japanese. Also use when the user provides a file path and asks to explain specific English sections, or when they want to understand English code comments, error messages, config files, or technical documentation.
29