spex-plan

Orchestrate a research-plan-execute cycle for feature development with traceability.

2|Updated Mar 8, 2026
One-click install
npx skills add https://github.com/ran729/context-rot-skill --skill spex-plan
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: spex-plan
Source: https://github.com/ran729/context-rot-skill/tree/main/.claude/.claude/skills/spex-plan
Command: npx skills add https://github.com/ran729/context-rot-skill --skill spex-plan

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill enforces a rigorous research-plan-execute cycle for feature development, ensuring complete traceability and preventing context rot by grounding AI actions in existing code, past requirements, and established policies.

Core Features & Use Cases

  • Ambiguity Resolution: Identifies and resolves ambiguities in user requests before planning.
  • Conflict Detection: Checks for semantic conflicts with existing code, policies, and past decisions.
  • Traceable Planning: Generates detailed plans with clear requirements, decisions, and impact analyses.
  • State Management: Orchestrates a strict state machine (INIT → RESEARCH → GENERATE_PLAN → REVIEWING_PLAN → COMPILING_TASKS → EXECUTE → COMPLETE) for feature development.
  • Use Case: When a user requests a new feature, this Skill guides the AI through a structured process to understand the requirements, check for conflicts, generate a detailed plan, and then execute it, ensuring all steps are documented and traceable.

Quick Start

Use the spex plan skill to start planning a new feature.

Frequently Asked Questions about spex-plan

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

FAQPage Schema
How do I prevent context rot when using AI for feature development?

To prevent context rot during feature development, you can orchestrate a strict research-plan-execute cycle that grounds AI actions in existing code, past requirements, and established policies to maintain full traceability.

How do I identify semantic conflicts before planning a new feature?

You identify semantic conflicts by checking new feature requests against existing code, established policies, and past decisions before generating a plan, ensuring the development workflow resolves ambiguities early.

What is the best way to enforce traceability in an AI-driven development workflow?

The best way to enforce traceability in an AI-driven development workflow is using a strict state machine that transitions through research, plan generation, and execution while documenting all requirements and impact analyses.

How do I resolve ambiguities in software requirements before generating an execution plan?

You resolve ambiguities in software requirements by identifying unclear user requests during the research phase and addressing them before transitioning to the plan generation state of the development cycle.

Can I use a state machine to manage feature development from research to execution?

Yes, you can use a strict state machine managing states like INIT, RESEARCH, GENERATE_PLAN, REVIEWING_PLAN, COMPILING_TASKS, EXECUTE, and COMPLETE to orchestrate the entire feature development lifecycle.

What are the limitations of AI orchestration without strict state management?

Without strict state management, AI orchestration suffers from context rot and loses traceability, meaning AI actions may drift from established policies and fail to document impact analyses accurately.