mspec-delta
Generates structured specification changes for project requirements using EARS format and automatic numbering.
Install
mkdir -p .claude/skills/mspec-delta && curl -L -o skill.zip "https://agentskills.codes/api/skills/download/12022" && unzip -o skill.zip -d .claude/skills/mspec-delta && rm skill.zipInstalls to .claude/skills/mspec-delta
Activation
This is the description your AI agent reads to decide when to run this skill — the better it matches your request, the more reliably it fires.
delta step of mspec workflow — generate Delta Spec with auto-numbered FR-NNNKey capabilities
- →Extract capability names from proposal files
- →Generate Delta Spec files with auto-numbered FR-NNN
- →Edit generated spec files with EARS format requirements
- →Add scenario examples with GIVEN/WHEN/THEN bullets
- →Validate Delta Spec for uniqueness and scenario structure
How it works
This skill extracts capabilities from a proposal, generates Delta Spec files with auto-numbered requirements, and guides the user to edit them in EARS format. It then validates the generated specs.
Inputs & outputs
When to use mspec-delta
- →Generate delta spec files
- →Format requirement documentation
- →Automate mspec requirement updates
About this skill
Procedure
- Run
mspec status --change <change-dir> --jsonfirst. - Read
proposal.md→ extract capability names from## Capabilities (touched). - For each capability, run
mspec delta init --capability <name> --change <change-dir>. The CLI auto-numbers FR-NNN by reading existingspecs/<name>/spec.md. - Edit each generated
changes/<change-dir>/specs/<capability>/spec.md:- Replace placeholders in
### Requirement: FR-NNN — <Short Title>headers (H3). - Write the Requirement body in EARS format using RFC 2119 keywords with the following semantics (per Constitution):
SHALL— 機能要件（system の振る舞いを規定するデフォルト）MUST— 制約・安全要件（絶対的強制または禁止）SHOULD— 推奨（逸脱可能だが理由が必要）MAY— 任意（オプション機能）
- EARS pattern guidance — read
"locale"from themspec status --jsonoutput (step 1) and use the matching format:- locale=ja の場合:
- Ubiquitous:
このシステムは SHALL <振る舞い>. - Event-Driven:
<トリガー> のとき、このシステムは SHALL <振る舞い>. - State-Driven:
<状態> の間、このシステムは SHALL <振る舞い>. - Unwanted Behavior:
<条件> の場合、このシステムは SHALL <振る舞い>. - Optional Feature:
<機能> が有効な場合、このシステムは SHALL <振る舞い>.
- Ubiquitous:
- locale=en の場合（locale フィールドが undefined の場合も含む）:
- Ubiquitous:
The system SHALL <response>. - Event-Driven:
When <trigger>, the system SHALL <response>. - State-Driven:
While <state>, the system SHALL <response>. - Unwanted Behavior:
If <condition>, then the system SHALL <response>. - Optional Feature:
Where <feature>, the system SHALL <response>.
- Ubiquitous:
- locale=ja の場合:
- Add at least one
#### Scenario: <Name>(must be H4) with- GIVEN,- WHEN,- THENbullets nested inside each Requirement.
- Replace placeholders in
readme.mdの## Artifacts節の- [ ] specs/*/spec.md (Delta Spec)を- [x] specs/*/spec.md (Delta Spec)に更新する。- Run
mspec validate --change <change-dir>(validates FR-ID uniqueness and H4 Scenarios). validate が失敗した場合は- [x] specs/*/spec.md (Delta Spec)を- [ ] specs/*/spec.md (Delta Spec)にロールバックする。 block: false— auto-continue via/mspec:continue.
Verification (C2)
mspec validate --change <change>— アーティファクト整合性チェックmspec anchor check --change <change>— アンカー解決確認mspec spec lint— Delta Spec実装詳細漏洩チェック
Learning (C3)
このスキルの実行で発生した学習候補を記録する:
<!-- LEARNING: <パターン説明> | source: <FR-ID> | confidence: low|medium|high -->
mspec learn コマンドが archive 済み changes からこれらを収集してpost-condition候補をproposeする。
When not to use it
- →When not generating a Delta Spec
- →When not using auto-numbered FR-NNN requirements
- →When not following EARS format
Limitations
- →Requires manual editing of generated spec files
- →Validation only checks FR-ID uniqueness and H4 Scenarios
- →Does not automatically write the requirement body
How it compares
This process automates the numbering and formatting of requirements into EARS format, ensuring consistency across spec documents.
Compared to similar skills
mspec-delta side by side with the closest alternatives in the catalog.
| Skill | Installs | Updated | Safety | Difficulty |
|---|---|---|---|---|
| mspec-delta (this skill) | 0 | 2mo | No flags | Intermediate |
| update-ai-context | 0 | 1mo | Review | Intermediate |
| writing-plans | 0 | 1mo | Review | Intermediate |
| korean-skill-creator | 8 | 8mo | Review | Beginner |
Try saying
Example prompts that trigger this skill in your AI assistant.
You might also like
update-ai-context
nuxeo
Update all AI context markdown files in the Nuxeo Web UI project to keep
writing-plans
AntonyKervazoCanut
Convert a spec or requirement into a numbered, executable implementation plan
korean-skill-creator
clwmfksek
한글 기반 클로드 스킬 자동 생성 도구. 사용자가 "클로드 스킬을 만들어줘" 또는 "[요구사항] 스킬 만들어줘"라고 요청할 때 사용. Progressive disclosure 원칙을 따르는 한글 문서 구조(SKILL.md + references/)를 자동으로 생성하고, 실전 예시를 포함한 일관성 있는 스킬 템플릿을 제공.
skill-forge
WilliamSaysX
Automated skill creation workshop with intelligent source detection, smart path management, and end-to-end workflow automation. This skill should be used when users want to create a new skill or convert external resources (GitHub repositories, online documentation, or local directories) into a skill. Automatically fetches, organizes, and packages skills with proactive cleanup management.
svelte-expert
Raudbjorn
Expert Svelte/SvelteKit development assistant for building components, utilities, and applications. Use when creating Svelte components, SvelteKit applications, implementing reactive patterns, handling state management, working with stores, transitions, animations, or any Svelte/SvelteKit development task. Includes comprehensive documentation access, code validation with svelte-autofixer, and playground link generation.
readme-generator
aws-samples
This skill should be used when users want to create or improve README.md files for their projects. It generates professional documentation following the Deep Insight/Strands SDK style - comprehensive yet focused, with clear structure and practical examples.