Spec to plan
Skill eugenepyvovarov/mcpbundler-agent-skills-marketplace/spec-to-plan
agent skills marketplace for mcp bundler(can be used with other ai tools with support of agent skills markeplaces). some skills are mine, also borrowed skills(credited to authors)
npx -y skills add eugenepyvovarov/mcpbundler-agent-skills-marketplace --skill spec-to-planAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 12 stars12 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
What its author says it does
Copied from the file, not written here
Transform project descriptions and feature requests into comprehensive specifications and actionable task lists. Use when the user wants to: (1) Create a specification from a project/feature description, (2) Generate a detailed plan with task breakdown, (3) Clarify requirements before implementation, (4) Convert ideas into structured development plans with progress tracking. Works with or without existing codebase files.
SKILL.md
4.6 KB, 844 tokens by cl100k_base, as published. Nobody here has run it
Spec-to-Plan Skill
This skill guides the process of transforming project descriptions into clear specifications and actionable task lists for developers.
Process Overview
The skill follows a five-phase workflow:
- Create Initial Specification - Analyze the feature/project description
- Clarify Questions - Identify and resolve all ambiguities
- Generate Markdown Spec - Produce detailed specification document (presented as
spec.mdin chat) - Create Comprehensive Todo - Break down into implementable tasks
- Output Plan Markdown - Generate final
plan.mdwith progress tracking (presented in chat)
Final deliverables: Both specification and plan content are provided as markdown in the conversation.
Phase 1: Initial Specification
Load the prompt template from references/phase1_spec_prompt.md and fill in the feature description provided by the user.
Key principles:
- If files are provided, analyze the codebase thoroughly
- If no files, work from the task description or ask for detailed project description
- Identify dependencies, structure, and integration points
- Note edge cases (within reason - ask about scope depth)
- List ALL unclear or ambiguous points
Do NOT implement yet - only analyze and prepare questions.
Phase 2: Clarify Questions
Ask all clarification questions at once. After receiving answers:
- If more questions persist, ask them
- Continue until no ambiguities remain
- Do NOT assume any requirements beyond explicitly described details
Scope control:
- Ask the user about desired depth and scope when uncertain
- Avoid going overboard with edge cases unless specifically requested
- Focus on what's explicitly needed
Phase 3: Generate Markdown Spec
After all questions are answered, create a comprehensive specification document and present it in the chat:
Content requirements:
- No specific template required (flexible format)
- Should be clear, detailed, and implementation-ready
- Include all decisions and clarifications from the Q&A phase
- Document the full feature/project specification
Output:
- Present the complete
spec.mdcontent directly in the chat - Use a clear header to indicate this is the specification document
Phase 4: Create Comprehensive Todo
Load the prompt template from references/phase2_plan_prompt.md.
Generate a detailed task breakdown that includes:
- Status tracking with emojis (🟩 Done, 🟨 In Progress, 🟥 To Do)
- Overall progress percentage at the top
- Confirmed decisions section
- Modular, minimal steps
- Important implementation details
- File-level changes (Add/Modify/Keep sections)
- Progress calculations
Critical: Do NOT add extra scope or unnecessary complexity beyond what was explicitly clarified.
Phase 5: Output Plan Markdown
Create the final plan.md content and present it in the chat:
Content requirements:
- Clear structure following the template in
phase2_plan_prompt.md - Checkbox tasks with status emojis
- Dynamic progress tracking
- Implementation details section
- File changes section listing specific files and modifications needed
Output:
- Present the complete
plan.mdcontent directly in the chat - Use a clear header to indicate this is the implementation plan
Both the specification and plan are provided as markdown content in the conversation.
Workflow Tips
When to analyze codebase:
- If user provides files → analyze them thoroughly
- If no files → work from description or ask for project details
Question iteration:
- Ask comprehensive questions in each round
- Wait for all answers before proceeding
- Continue asking until complete clarity is achieved
Scope management:
- Always ask about desired depth when uncertain
- Keep steps elegant and minimal
- Ensure seamless integration with existing codebase
Output quality:
- Plans should lead to simple, elegant, minimal code
- No extra complexity beyond discussed requirements
- All steps should be modular and implementable
Reference Files
references/phase1_spec_prompt.md- Initial specification prompt templatereferences/phase2_plan_prompt.md- Plan generation prompt with examples
What ships with it: 3 files
3.5 KB alongside SKILL.md
.mcp-bundler/
- manifest.json282 B
references/
- phase1_spec_prompt.md1.1 KB
- phase2_plan_prompt.md2.2 KB