speckit-plan

Name: speckit-plan
Availability: InStock
Author: tykhoncode

Generates implementation plans from feature specs with resolved unknowns and contract artifacts.

Updated May 3, 2026

Search Tags:#implementation planning #spec automation #research synthesis #extension hooks #agent context update #architecture contracts

Ready to run

npx skills add https://github.com/tykhoncode/dicekit --skill speckit-plan-tykhoncode

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: speckit-plan
Source: https://github.com/tykhoncode/dicekit/tree/main/.claude/skills/speckit-plan
Command: npx skills add https://github.com/tykhoncode/dicekit --skill speckit-plan-tykhoncode

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It helps turn an input feature spec into a structured implementation plan with resolved unknowns, consistent design artifacts, and automated extension hooks when available.

Core Features & Use Cases

Unknowns-to-research workflow: Converts NEEDS CLARIFICATION items into targeted research tasks and consolidates results into research.md.
Design & artifact generation: Produces data-model.md, contract docs under contracts/, and quickstart.md, then updates agent context markers in CLAUDE.md.
Extension hook integration: Detects optional and mandatory hooks from .specify/extensions.yml (before and after planning) and prints the appropriate execution directives.

Quick Start

Ask for the implementation planning by running the speckit-plan workflow with optional guidance as your input, and it will generate research.md, data-model.md, contracts/, and quickstart.md while reporting the plan path and produced artifacts.

Frequently Asked Questions about speckit-plan

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I generate an implementation plan from a feature spec?▼

To generate an implementation plan from a feature spec, the workflow resolves NEEDS CLARIFICATION items via targeted research, then produces data-model.md, contract docs, and quickstart.md artifacts. It enforces gate checks and updates agent context markers in CLAUDE.md.

What is the best way to resolve unknowns in a feature spec before planning?▼

The best way to resolve unknowns in a feature spec is converting NEEDS CLARIFICATION items into targeted research tasks. The workflow consolidates findings into research.md to eliminate ambiguity before producing architecture contracts and data models.

Do I need a specific directory structure to automate spec planning?▼

Yes, spec automation requires a .specify directory structure. You must run the setup-plan PowerShell script, load constitution memory, and follow the IMPL_PLAN template phases to successfully generate the planning artifacts.

Can I use custom extension hooks during the implementation planning workflow?▼

Yes, you can use custom extension hooks during implementation planning by defining optional and mandatory hooks in .specify/extensions.yml. The workflow detects before_plan and after_plan hooks and prints appropriate execution directives.

What artifacts are produced when creating architecture contracts from a feature spec?▼

Creating architecture contracts from a feature spec produces research.md, data-model.md, contract documents under the contracts/ directory, and quickstart.md. It also updates agent context markers within CLAUDE.md.

How does the workflow update agent context after generating a plan?▼

The workflow updates agent context by modifying markers in CLAUDE.md after generating design artifacts. This ensures the agent operates with the latest data-model and contract definitions during subsequent implementation phases.